APPROVED: C. Reid Ferring, Major Professor and Chair of

the Department of Geography Lisa Nagaoka, Minor Professor Miguel Acevedo, Committee Member Sandra L. Terrell, Dean of the Robert B.

Toulouse School of Graduate Studies

INTERPRETING PREHISTORIC PATTERNS: SITE-CATCHMENT ANALYSIS IN THE

UPPER TRINITY RIVER BASIN OF NORTH CENTRAL TEXAS

Marikka Lin Williams, B.F.A.

Thesis Prepared for the Degree of

MASTER OF SCIENCE

UNIVERSITY OF NORTH TEXAS

December 2004

Williams, Marikka Lin, Interpreting Prehistoric Patterns: Site Catchment Analysis

in the Upper Trinity River Basin of North Central Texas. Master of Science (Applied

Geography), December 2004, 301 pp., 25 tables, 92 illustrations, references, 93 titles.

Archaeologically site-catchment analysis produces valuable information regarding

prehistoric subsistence strategies and social organization. Incorporating archaeological

data into catchment analyses is an effective strategy to develop regional models of

prehistoric site selection and settlement patterns. Digital access to data permits the

incorporation of multiple layers of information into the process of synthesizing regional

archaeology and interpreting corresponding spatial patterning. GIS software provides a

means to integrate digital environmental and archaeological data into an effective tool.

Resultant environmental archaeology maps facilitate interpretive analysis. To fulfill the

objectives of this thesis, GIS software is employed to construct site-catchment areas for

archaeological sites and to implement multivariate statistical analyses of physical and

biological attributes of catchments in correlation with assemblage data from sites.

Guided by ecological, anthropological and geographical theories hypotheses testing

evaluates patterns of prehistoric socio-economic behavior. Analytical results are

summarized in a model of prehistoric settlement patterns in North Central Texas.

ii

Copyright 2004

by

Marikka Lin Williams

ACKNOWLEDGEMENTS I am very thankful to many people who were instrumental in the process of

completing my thesis. First and foremost, I would like to thank God, my parents, my

siblings, my friends and a number of musicians who have provided continuous

emotional, as well as, spiritual support, throughout the years. Thank you with my

eternal love and respect.

I am also very thankful to all of my teachers who have provided academic support

throughout my education. While I cannot thank all who touched my life and enriched

my education there are several that I would like to mention who have been particularly

influential. Thanks to the first schoolteacher who made an impact on my life during

my early childhood, Mrs. Reckon. Thanks to Dr. Manning from High School who helped

me to open my mind and expand my horizons. Thanks to Dr. Kim and Dr. Beason from

Stephen F. Austin State University. Thanks to Dr. Lindsey and Marilyn Patton, from the

Willis Library, who have been supportive, throughout my education at the University of

North Texas. Thanks to the people from the UNT Registrars office and the Financial Aid

Office, who kept me on board after I graduated from UNT with my Bachelor’s Degree.

Thanks to Ellen Turner, Dr. Thomas Hester and Rowman and Littlefield Publishers who

granted permission to use projectile point illustrations from A Field Guide to Stone

Artifacts of Texas Indians. Finally, thanks to the UNT Geography Department including

Dr. Lyons, Tami Deaton, Dr. Schoolmaster, Bruce Hunter, Dr. Williams, Dr. Acevedo, Dr.

Nagaoka and Dr. Ferring. Thanks to all of the people who have contributed their

knowledge toward completion of this research. You are immensely appreciated.

iii

TABLE OF CONTENTS ACKNOWLEDGMENTS iii LIST OF TABLES viii LIST OF FIGURES x CHAPTER 1: INTRODUCTION 1 Research Orientation 2 Objectives 15 CHAPTER 2: METHODOLOGICAL FOUNDATIONS 18 Cultural Ecology 18 Geographical Theory 22 Site Catchment Analysis 24 Site Catchment Analysis Case Studies 29 CHAPTER 3: RESEARCH SETTING 36 Geology of North-Central Texas 36 Topography 39 Drainage Systems 39 Physiographic Setting 42 North-Central Texas Soils 46 Floral Resources in North-Central Texas 53 Faunal Resources in North-Central Texas 61 Climate of North-Central Texas 71 Past Environments 75

iv

CHAPTER 4: CULTURAL CHRONOLOGY OF NORTH CENTRAL TEXAS 77 Paleoindian (11,500-8,500 BP) 83 Paleoindian Settlement Patterns (11,500-8,500BP) 88 Archaic (8,500-1,250 BP) 89 Early Archaic (8,500-6,000 BP) 93 Middle Archaic (6,000-3,500 BP) 96 Late Archaic (3,500-1,250 BP) 99 Archaic Climatic Change (8,500-1,250 BP) 103 Archaic Subsistence Patterns (8,500-1,250 BP) 104 Early Archaic Settlement Patterns (8500-6000BP) 105 Middle Archaic Settlement Patterns (6000-3500BP) 106 Late Archaic Settlement Patterns (3500-1250BP) 107 Late Prehistoric (1250-250BP) 108 Late Prehistoric Climatic Conditions 113 Late Prehistoric Subsistence Patterns 113 Late Prehistoric Settlement Patterns 115 Unanswered Questions 116 CHAPTER 5: RESEARCH METHODOLOGY 118 Soils as a Diagnostic Tool 124 Soils Variables 125 Soil Settings 126 Soil Potential for Plants and Wildlife 129

v

Interpretations: Potential Economic Value of Soils 130 Site Catchment Soils Diversity 136 Geological Variables 137 Physiographic Variables 139 Climatic Variables 142 Archaeological Variables 144 Site Catchment Analysis of Prehistoric Sites in North-Central Texas 154 Hypotheses Testing 158 Study Area 165 CHAPTER 6: RESULTS 168 Geological Setting 168 Physiographic Setting 173 Topographic Setting 178 Soils: Great Groups 183 Soils: Soil Settings 189 Soils: Economic Interpretations 191 Site Catchment Diversity 204 Climatic Interpretations 206 Site Activities and Density 208 Statistical Conclusions 216 CHAPTER 7: NORTH CENTRAL TEXAS SETTLEMENT PATTERNS 221 Prehistoric Populations along the West Fork of the Trinity 221

vi

Prehistoric Populations along the Elm Fork of the Trinity 227 Prehistoric Populations along the East Fork of the Trinity 235 Functional Site Types in North-Central Texas 238 Prehistoric Adaptations 240 CHAPTER 8: DISCUSSION AND FUTURE RECOMMENDATIONS 246 Discussion 246 Future Recommendations 248 APPENDIX 253 GIS GLOSSARY 293 REFERENCES 294

vii

LIST OF TABLES Table 1 Various chronologies developed for cultural subdivisions 80 Table 2 Soil taxonomy classifications in study area 128 Table 3 Weighting procedure for plant and wildlife estimates 129 Table 4 Geological formations associated with site catchment areas 138 Table 5 Site catchment hypotheses 162 Table 6 Site function hypotheses 163 Table 7 Economic values hypotheses 164 Table 8 Statistical results summary 217 Table 9 Statistical results summary 218 Table 10 Statistical results summary 219 Table A1 Raw data for hypothesis testing 278 Table A2 Raw data for hypothesis testing 279 Table A3 Raw data for hypothesis testing 280 Table A4 Raw data for hypothesis testing 281 Table A5 Raw data for hypothesis testing 282 Table A6 Raw data for hypothesis testing 283 Table A7 Raw data for hypothesis testing 284 Table A8 Raw data for hypothesis testing 285 Table A9 Economic values: Vegetation scores for soil settings 286 Table A10 Economic values: Vegetation scores for soil settings 287

viii

Table A11 Economic values: Vegetation scores for soil settings 288 Table A12 Economic values: Vegetation scores for soil settings 289 Table A13 Economic potential of site catchment areas 290 Table A14 Vegetation potential of site catchment areas 291 Table A15 Wildlife potential of site catchment areas 292

ix

LIST OF FIGURES Figure 1 Lakes and major cities in north-central Texas 3 Figure 2 Geology of north-central Texas 38 Figure 3 Hydrological networks of north-central Texas 40 Figure 4 Physiographic regions of Texas and north-central Texas 43 Figure 5 Soil distribution in north-central Texas study area 47 Figure 6 Archaeological faunal distribution from excavated sites in study area 67 Figure 6a Fauna recovered from archaeological sites in study area 67 Figure 6b Archaeological fauna and change through time 68 Figure 6c Archaeological fauna and change through time 69 Figure 7 Precipitation patterns in Texas and north-central Texas 72 Figure 8 Average monthly temperature and precipitation trends in north Texas 73 Figure 9 Average monthly temperature and precipitation trends comparisons 74 Figure 10 Distribution of prehistoric sites in Texas and north-central Texas 78 Figure 11 Generalized projectile point chronology in north-central Texas 81 Figure 12 Chronological overview of projectile point types 82 Figure 13 Distribution of prehistoric sites in Texas and north Texas 84 Figure 14 Paleoindian projectile point chronology 86 Figure 15 Distribution of Archaic sites 90 Figure 16 Archaic projectile points in north-central Texas 92 Figure 17 Distribution of Early Archaic sites 94 Figure 18 Early Archaic projectile points 95

x

Figure 19 Distribution of Middle Archaic sites 97 Figure 20 Middle Archaic projectile points 98 Figure 21 Distribution of Late Archaic sites 101 Figure 22 Late Archaic projectile points 102 Figure 23 Distribution of Late Prehistoric sites 109 Figure 24 Late Prehistoric projectile points 111 Figure 25 Conceptual diagram of data included in site catchment analysis 122 Figure 26 Procedure to link economic value to site catchment 135 Figure 27 Relative abundance of plant family in study area from Ferndale Bog 143 Figure 28 Diagram of GIS data processing design in site catchment analysis 155 Figure 29 Conceptual diagram of GIS in relation to components of the research 156 Figure 30 North central Texas research study area 166 Figure 31 North central Texas research study area geology 170 Figure 32 Geological settings and time periods 171 Figure 33 Geological settings and occupation frequency 172 Figure 34 Physiographic settings in north-central Texas study area 175 Figure 35 Physiographic settings and time periods 176 Figure 36 Physiographic settings and occupation frequency 177 Figure 37 Topographic settings and time periods 179 Figure 38 Topographic settings and occupation frequency 180 Figure 39 Elm Fork site catchment elevations 181 Figure 40 West Fork site catchment elevations 182

xi

Figure 41 Elm and West Fork great group distribution and site catchments 184 Figure 42 East Fork great group distribution and site catchments 185 Figure 43 Great groups and time periods 186 Figure 44 Great groups and occupation frequency 188 Figure 45 Soil settings and time periods 189 Figure 46 Soil settings and occupations frequency 190 Figure 47 Economic values and prehistoric components 192 Figure 48 Economic values and occupation frequency 194 Figure 49 Economic values and time periods 195 Figure 50 Vegetation potential and prehistoric components 197 Figure 51 Vegetation potential and occupation frequency 198 Figure 52 Vegetation potential and time periods 199 Figure 53 Wildlife potential and prehistoric components 201 Figure 54 Wildlife potential and occupation frequency 202 Figure 55 Wildlife potential and time periods 203 Figure 56 ANOVA of site catchment diversity and occupation frequency 204 Figure 57 ANOVA of site catchment diversity and change through time 205 Figure 58a Distribution of climatically abundant vegetation families 206 Figure 58b Distribution of climatically abundant vegetation families 207 Figure 59 Prehistoric activities and time periods 209 Figure 60 Prehistoric activities and occupation frequency 210 Figure 61 Prehistoric activities, site density and geological settings 211

xii

Figure 62 Prehistoric activities, site density and soil settings 212 Figure 63 Prehistoric activities, site density and great groups 213 Figure 64 Prehistoric activities, site density and topographic settings 214 Figure 65 Prehistoric activities, site density and physiographic settings 215 Figure 66 Prehistoric populations along the West Fork of the Trinity 226 Figure 67 Prehistoric populations along the Elm Fork of the Trinity 234 Figure 68 Prehistoric populations along the East Fork of the Trinity 237 Figure A1 Wildlife potentials and Prehistoric sites 254 Figure A2 Wildlife potentials and Archaic projectile points 255 Figure A3 Wildlife potentials and Early Archaic projectile points 256 Figure A4 Wildlife potentials and Middle Archaic projectile points 257 Figure A5 Wildlife potentials and Late Archaic projectile points 258 Figure A6 Wildlife potentials and Late Prehistoric projectile points 259 Figure A7 Wildlife potentials and Late Prehistoric I projectile points 260 Figure A8 Wildlife potentials and Late Prehistoric II projectile points 261 Figure A9 Vegetation potentials and Prehistoric sites 262 Figure A10 Vegetation potentials and Archaic projectile points 263 Figure A11 Vegetation potentials and Early Archaic projectile points 264 Figure A12 Vegetation potentials and Middle Archaic projectile points 265 Figure A13 Vegetation potentials and Late Archaic projectile points 266 Figure A14 Vegetation potentials and Late Prehistoric projectile points 267 Figure A15 Vegetation potentials and Late Prehistoric I projectile points 268

xiii

Figure A16 Vegetation potentials and Late Prehistoric II projectile points 269 Figure A17 Economic potentials and Prehistoric Sites 270 Figure A18 Economic potentials and Archaic projectile points 271 Figure A19 Economic potentials and Early Archaic projectile points 272 Figure A20 Economic potentials and Middle Archaic projectile points 273 Figure A21 Economic potentials and Late Archaic projectile points 274 Figure A22 Economic potentials and Late Prehistoric projectile points 275 Figure A23 Economic potentials and Late Prehistoric I projectile points 276 Figure A24 Economic potentials and Late Prehistoric II projectile points 277

xiv

CHAPTER 1

INTRODUCTION

Archaeological sites are part of a human ecosystem built within a dynamic living

physical landscape. It is within this interactive matrix that prehistoric communities

interacted spatially, socially and economically. The nature of this matrix is typically

extrapolated from subsistence activities and settlement patterns. Evidence of

subsistence activities is primarily recovered from archaeological sites in the form of

fauna (i.e. animal remains), features (e.g. fire-cracked rock, hearths, middens) and

artifacts (e.g. projectile points, other stone tools, grinding stones, ceramics). In order

to facilitate pattern recognition and assist in the modeling of prehistoric settlement

patterns, archaeologists have examined the characteristics of the micro-environments

that surround site locations. However, broad descriptive generalizations based on

macro-environmental correlations are often found in the literature,

The study of prehistoric sites is not usually complemented by some treatment of their setting, the situation in their immediate vicinity—the principal concern of the inhabitants—tends to be neglected or at any rate overshadowed by generalized statements regarding the physiographic, vegetational, climatic or kindred zones of which they form a part (Vita-Finzi and Higgs, 1970:1).

This is particularly the case for the state archaeological site files housed at the Texas

Archaeological Resource Laboratory (TARL) in Austin. One of the most significant

components missing from state archaeological site files is the specific environmental

context within which these sites and artifacts were discovered. Records housed in state

and university laboratories are increasing in number, yet often site report content is not

substantial enough to facilitate comprehensive regional studies. Glimpses into

1

prehistory that have been provided by archaeological discoveries must be viewed

cumulatively in order to understand the larger patterns that emerge regarding

prehistoric behavior. Although many contract, academic and published reports from the

region make reference to soils, vegetation, geology and zonal physiographic

characteristics of the landscape, there is a general disconnection between the sites, the

data and the interpretive inter-relationships. This research seeks to bridge the gap by

providing a series of maps with interpretive visual representations that provide a bird’s

eye view and facilitate pattern recognition toward developing a more comprehensive

view of site distribution within the environment. Tabular data coupled with interpretive

discussion based on spatial and statistical analysis is used to explain these patterns.

Research Orientation

A substantial amount of research has been conducted, in north-central Texas, with

respect to prehistoric socio-economic systems (Dawson and Sullivan, 1969; Hays et al.,

1972; Bousman and Verrett, 1973; Nunley, 1973; McCormick et al., 1975, Filson et al.,

1975; McCormick, 1976; Lynott, 1977, 1981; Skinner and Baird, 1985; Lebo and Brown,

1990; Prikryl, 1990; Ferring and Yates, 1986, 1997, 1998). These archaeological reports

involving cultural-ecology, subsistence and settlement pattern studies have provided a

wealth of empirical data. Much of the contract and academic research in this region,

conducted between 1968 and 1998, has resulted in the development of models that

discuss site function, distribution and regional settlement patterns. It is important to

note that the research conducted in north-central Texas (figure 1) has revolved around

contract work associated with the construction of reservoirs. Along the Elm Fork of the

2

Trinity River extensive professional archaeological investigations have been conducted.

Research along the East Fork is based on academic and professional efforts but to a

much lesser extent than the Elm Fork. As a result, there is a strong bias in site

distribution related to the intensity of investigations along the Elm Fork. West Fork

sites were primarily reported by amateur archaeologists. The lack of professional

investigations along the West Fork partially explains the paucity of sites. However, the

sites that do exist offer valuable comparative environmental data for site catchment

analysis (SCA).

Figure 1. Lakes and major cities in north-central Texas.

3

Although a variety of regional models have been proposed for this “region”, many of

these models are built in relation to specific excavation sequences typically located in

isolated microenvironments, which are considered to be representative of the region.

General physiographic characteristics are frequently employed to fill in the gaps in order

to extend the model beyond the immediate study area.

Previous studies conducted along the Elm Fork of the Trinity River have resulted in a

variety of interpretations of prehistoric settlement patterns. This section will review the

settlement pattern theories that have been formulated based on prior research in the

region. Several investigators have referred to this region as "marginal" (Lynott, 1977;

Skinner and Baird, 1985; Prikryl, 1990; Story, 1990). However, other investigators such

as Dawson and Sullivan (1969: 5) consider the grassland, woodland, riverine and

transitional ecotonal settings to be “highly opportune” for prehistoric occupation.

Dawson and Sullivan acknowledge the economic potential of the variety of resources

within each environment. Within the grassland, a variety of vegetation and faunal

resources are accessible. Riverine environments, nested within the woodlands, provide

a substantial supply of protein associated with fishing and gathering activities.

Throughout the entire region, small fauna, supplementary foods, medicine, lithics, wood

and various sources for construction can be obtained (Dawson and Sullivan, 1969).

Since multiple environments are present in north-central Texas these authors speculate

that prehistoric groups utilized a generalized and vast array of resources. A highly

mobile and semi-sedentary model of occupation is posited within all three settings.

Based on the character of the environment they propose that food could be obtained

4

during extremely dry or wet climates if groups remained small and spread out across

the landscape. A scattered settlement pattern would sustain balance with climatic and

seasonal fluctuations. Differences in site positioning between Archaic ridge sites and

bottomland Late Prehistoric sites, on the East Fork of the Trinity, are considered to be a

intentional selection of site location geared towards maintaining this balance (Dawson

and Sullivan, 1969).

A food surplus, stored during the spring and summer, would probably be required for group survival during the late winter and early spring. At that time, the probable high water, hibernation of many woodland animals and the scattering of grassland herds would vastly reduce food production. The usual size of the surplus would probably determine functional group size. This surplus could be achieved by hunting and gathering groups from herd slaughter, grass seeds and gathered woodland and lowland products (Dawson and Sullivan, 1969: 6).

Fundamental assumptions of their settlement pattern theory are that “site positioning

reflects basic needs that vary according to cultural adjustment; and that sites are

located near the major source of food” (Dawson and Sullivan, 1969: 2). Given that

several food sources are available, location is thought to have been based on situating

women and children near areas where gathered food sources are maximized (Dawson

and Sullivan, 1969). Essentially, Dawson and Sullivan conclude that prehistoric

population size was primarily dependent on the environment in correlation with the

seasons and the climate. As a result, they propose that an environmentally based

economic threshold was established that dictated prehistoric group size, site selection

and organization.

Hays et al. (1972) identify three distinctive geomorphic locations related to

archaeological sites in an environmental impact study conducted along the Elm Fork of

5

the Trinity. Floodplains, terraces, and uplands are considered to be preferred locations

for archaeological sites (Hays et al., 1972). Floodplain sites investigated by Hays and

others primarily consist of concentrations of burned rock, mussel shells, flint tools and

debitage. According to these investigators several floodplain sites seem to represent

single events of preparing and eating mussels. “Projectile points are present and

indicate this activity may have resulted from a small hunting party stopping for a meal

or for overnight” (Hays et al., 1972: 18). Terrace sites are reported to predominantly

contain lithic scatter and upland sites are typically found where natural quartzite and

chert deposits occur on the surface (Hays et al., 1972). Prehistoric populations

probably quarried raw material at upland locations given the association with local raw

materials (Hays et al., 1972). Faunal remains of both large and small species indicate

hunting in the floodplain and uplands. Remains from the sites, investigated during their

survey, indicate utilization of a variety of microenvironments (Hays et al., 1972).

Overall, Hayes categorizes sites as representing activities such as food preparation

and/or camping, short term seasonal occupations or long term base camps sites that

provided access to various environmental resources.

During an archaeological reconnaissance of the proposed Aubrey reservoir

Bousmann and Verrett (1973) define six micro-environmental zones on the basis of

correlated biology, geology and physiography. These zones include: “rivers and

drainages; floodplains; floodplain rises; fluvatile terraces; upland slopes; and uplands”

(Bousmann and Verrett, 1973: 7). Their purpose for developing these categories is to

provide an explanatory mechanism for inter- and intra-site variation (Bousmann and

6

Verrett, 1973). Employing this model, Bousmann and Verrett (1973: 12) hypothesize

that “the Eastern Cross Timbers district was utilized more often for base camps, and the

prairies were utilized for specialized activities, i.e. hunting.” Supporting evidence for

this hypothesis includes the “presence of permanent water sources, concentration of

wood, highly permeable sandy soils and a stable supply of native plants and animals in

the Cross Timbers” (Bousmann and Verrett, 1973: 12). Another hypothesis, formulated

by these investigators, is that fluvatile terraces specifically function most often as base

camp sites while other zones are occupied only when seasons or climatic conditions

permitted and provided sufficient returns (Bousmann and Verrett, 1973). The central

location of terraces maximized access to all of the resource zones and proximity to

permanent water sources in areas that would be protected from flooding (Bousmann

and Verrett, 1973). In relation to this supporting environmental evidence it is posited

that prehistoric groups moved to upland camps during wetter seasons and lowland sites

during drier seasons. Deviations from this model are expected due to variations in

social organization, religious factors, taboos and other socially defined parameters.

McCormick et al. (1975) proposes that this region served as an avenue of seasonal

migration due to the increased availability of eco-tonal resources. Based on a survey

project conducted along the southern reaches of the Elm Fork these investigators

observe that Archaic seasonal camps are typically located in eco-tonal settings and

secondary activity-specific sites are situated in relation to specific outcroppings, plants,

and water (McCormick et al., 1975: 41). McCormick (1975) posits that Late Prehistoric

groups followed similar patterns, producing seasonal camps and activity specific sites,

7

until agriculture was introduced and large base camps or permanent villages emerged.

According to McCormick’s settlement model, groups seasonally migrate with the bison

and return north in spring to spend the summer months in large agricultural villages.

During the late fall, family sized groups would leave the base camps to follow the bison herds south into the central Texas area, by following the Cross Timbers from the Red River south to the vicinity of Waco, Texas. Temporary camps, showing signs of both male and female activities, should be located well into the Cross Timbers and out on the prairies in association with whatever resource was being exploited. These will be fewer in number than during the Archaic period because of the transitory nature of the usage (McCormick et al., 1975: 41).

The proposed settlement model divides sites into Archaic and Neo-American (Late

Prehistoric) subdivisions. These subdivisions are classified into three groups:

permanent village, temporary campsites and activity-specific sites. In the study area,

located south of Lewisville, along the Elm Fork of the Trinity, this model is found to be

useful. As predicted the large village type settlement is also present in the study area.

Temporary campsites for both the Archaic and Neo-American periods are situated in

similar ecological microenvironments. Permanent villages and activity-specific sites,

such as manufacturing stations, are not recognized within their study area. However,

these site types are found to the southwest further into the Cross Timbers. According

to the model, temporary camps of Neo-American and Archaic groups should be located

in similar areas with affinities in site structure with the exception of diagnostic tools

such as large “dart” points associated with the Archaic and small "arrow" points with

ceramics which identify the Neo-American groups (McCormick et al., 1975: 45). All of

the Neo-American sites are ephemeral and situated along the upland edge, providing

immediate access to a maximum of resource zones. Seasonal Archaic sites tend to

8

display more “permanent characteristics” since these sites were apparently occupied for

a longer duration (McCormick et al., 1975: 45). McCormick and others hypothesize that

the zone of movement was along the prairie-cross timbers boundaries attributed to the

presence of more “permanent” sites (Prikryl, 1990: 31). McCormick et al. (1975) find

Lewisville sites located respectively north and south in almost identical environmental

situations. What is considered to be unusual is the “apparent similarity of activity

regardless of temporal considerations” (McCormick et al., 1975: 78).

The amount of patterning in artifacts and activity areas within a site will increase in proportion to an increase in occupation time. Neo-American permanent villages will display the most evidence of "city planning", Archaic seasonal camps next, then Neo-American transitory camps, and lastly, both Neo-American and Archaic activity-specific stations. In the absence of diagnostic artifacts it is difficult to distinguish Archaic from Neo-American stations where similar activities took place (McCormick et al., 1975: 46).

Reoccupation of sites over a long period of time is also considered to be common in this

region. The settlement pattern exhibited here is thought to have been established

during the Paleoindian and continued until the “westward expansion of the Europeans

altered the native cultures to such an extent that they could no longer continue their

normal settlement and subsistence patterns” (McCormick et al., 1975: 78).

Lynott (1977) presents a regional model for future archaeological research in north-

central Texas and provides an extensive listing of previous investigations in the prairie-

cross timbers area of the region. Lynott (1977) emphasizes subsistence-settlement

models and encourages use of general systems theory. His regionally oriented design

hypothesizes changes in adaptive strategies through time. During the Early to Middle

Archaic, Lynott (1977) observes a shift from open-land to big game hunting to

9

bottomland riverine hunting and gathering. An “increase of population density and

decrease in territoriality” defines the Late Archaic (Lynott, 1977: 158). As a result,

Lynott (1977: 158) proposes a “restricted wandering” settlement pattern where sites

are situated in terraces settings near water and bottomland resources. During the first

phase of the Late Prehistoric, Lynott proposes that the initial stages of tribalization,

sedentism and horticulture were related to an increase in population pressure that

instigated territorial conflicts (Lynott, 1977: 159). The concept of tribalization is related

to the presence of ceremonial structures that emerged on the East Fork of the Trinity.

This situation is thought to have continued to a greater extent during the second phase

of the Late Prehistoric as the “population grew and social boundaries became more

defined” (Lynott, 1977: 160). Throughout time Lynott emphasizes the utilization of

bottomland resources. Regionally, he proposes use of the Wichita model of settlement

and subsistence patterns. This model is based upon a nomadic winter bison hunt in the

prairies with semi-permanent villages occupied during the spring and summer (Lynott,

1977). An alternative model is proposed by associations with the Tonkawa that

suggests seasonal hunting and gathering in north-central Texas (Lynott, 1977). In

general, the settlement pattern that Lynott (1977) proposed for sites revolves around

large bottomland base camps and smaller, seasonally occupied, special activity sites.

Skinner and Baird (1985: 5-11) propose a “marginal territory” model for the Ray

Roberts Lake area, along the northern branch of the Elm Fork. This model is based on

the paucity of sites, small site size, low density of cultural remains at individual sites

and numerous natural environmental variables. It operates on the premise that the

10

diverse resources in the region served as specialized ecosystems. In particular, the

mast resources provided in the uplands of the Cross Timbers and the riparian species,

documented by archaeological reports, reflect these specialized ecosystems. Field

observations of maintenance and extraction sites reveal a concentration in the Cross

Timbers and riparian environmental zones (Skinner and Baird, 1985). Conclusions

based on research conducted at these sites are made according to a demographic

version of a settlement model known as the central based wandering (CBW) community

pattern. The “CBW model”, developed by Beardsley and others in 1956, has been used

as an “organizational tool in various parts of Texas and surrounding regions” (Skinner

and Baird, 1985: 5-1). It is employed as a means of “categorizing sites which had a

limited number of surface artifacts” and where diagnostic tools or pottery are not found

(Skinner and Baird, 1985: 5-1). Due to the lack of datable artifacts, sites are

categorized into functional types on the basis of the character of lithic debris present on

the surface of sites and artifact assemblages. It was expected that by correlating these

two classes of information, consistent site types could be defined and could be

replicated by others (Skinner and Baird, 1985: 5-3).

Based on the results of the site survey, it is noted that 40 of the prehistoric sites

appear to represent a single occupation period while 22 have artifacts from more than

one occupation period. Twenty-one morphological site types are defined by cluster

analysis of the survey data and then consolidated into 7 functional sites (Skinner and

Baird, 1985). Results indicate that of the 117 sites surveyed 62 were datable, 23 were

base camps, 32 seasonal camps, 16 hunting camps, 10 musselling camps, 9 hunting

11

stations, 2 collecting stations and 21 lithic procurement sites (Skinner, 1985).

According to Skinner (1985), undated sites typically include locations where stone was

gathered, or quarried, and where initial tool making processes were carried out.

Skinner and Baird (1985) pay particular attention to site type, age, location, and

other variables such as assemblage, deposition and preservation. Skinner and Baird

(1985) also provide an alternative model of settlement that utilizes the work of others in

the surrounding areas for assistance in the reconstruction of the effective natural

environment. This alternative settlement model associates cultural change and shifts in

settlement locations with changes in the natural environment. It is noted that emerging

patterns indicate that certain locations were preferred over time. According to Skinner

and Baird (1985), while repeated reoccupation may distort functional differences, this

situation provides a means of observing the use of a particular site location during its

lifetime. However, besides reference to the mast crops, there is no specific discussion

as to why these locations were preferred and if there are patterns in particular

environmental variables that made them desirable locations for occupation. There is

also a lack of clarity as to the shifts in settlement locations that were reportedly

responding to environmental change.

A series of observations from the Lewisville report (Ferring and Yates, 1998) and

Ray Roberts report (Ferring and Yates, 1997) shed light on prehistoric settlement

patterns and provide informative data regarding site selection along central portion of

the Elm Fork of the Trinity. Site location patterns at Lake Lewisville indicate the

utilization of ecotonal resources associated with the transition between the prairies and

12

woodlands. Faunal data documented from archaeological excavations indicates that

prehistoric populations may have been focusing on hunting on the wooded edges rather

than harvesting the potential mast crops at central locations within the Cross Timbers

(Ferring and Yates, 1998). Accordingly sites are primarily situated on the eastern edge

of the Cross Timbers near the border of the Blackland Prairie.

Topographic settings are a fundamental component of site selection that is often

incorporated but not explained interpretively. For example, one could ask why the

upland and terraces settings would be preferred over floodplains. Of the 150

components in the Lewisville Lake area 49% are located in terrace settings, 49% in

upland settings and 2% in the floodplains (Ferring and Yates, 1998). With respect to

drainages 45% are located along Little Elm Creek, 27% along the Elm Fork of the

Trinity, 20% along Hickory Creek and 8% are situated along minor tributaries (Ferring

and Yates, 1998). Given that these sites are primarily located on terraces and upland

settings rather than in the floodplain may indicate that the Lake Lewisville area was

occupied during wetter climatic episodes. Artifact assemblages excavated in this area

suggest seasonal occupations so it may be that these areas were occupied during the

wetter seasons rather than wetter climates.

The overall pattern detected by Ferring and Yates (1998: 153) suggests that

prehistoric populations “exploited whatever was available to them, and possibly did so

within quite well-defined territories. Life was rigorous and large sedentary communities

were probably never common if they existed at all.” Ferring (1998: 153) concludes that

while evidence supports an increase in resource availability, stimulated by climate

13

change, in the Late Prehistoric, there is little evidence to support that “this stabilized

culture groups as indicated by decreased mobility, increased specialization of patch

exploitation, or decreased diversity of floral-faunal resource procurement practices.”

Despite all of the investigations that have been conducted in north-central Texas it

is the consensus of several archaeologists (Lynott, 1981; Skinner and Baird, 1985;

McCormick et al., 1975; Prikryl, 1990; Ferring, 1998) that information about subsistence

and settlement patterns of the prehistoric occupants of north-central Texas needs to be

expanded upon and models refined on a regional scale. According to Ferring (1998),

the absence of a local record of past environments and chronological controls has

prevented previous settlement models from consideration of diachronic adaptive

changes. As a result, regional settlement models have not actively considered changes

in past environments as possible controls on site frequencies or site characteristics.

Throughout the discussion, of the proposed regional models, there is no mention of

sites along the West Fork of the Trinity and typically Elm Fork investigations do not

incorporate East Fork research. Although these investigators are certainly familiar with

the cultural history of the region it appears that their “regional” settlement pattern

models are situated along and primarily in reference to either the Elm Fork or the East

Fork of the Trinity. In addition, although the settlement models refer to environmental

settings, the bulk of the analyses revolve around archaeological remains. While the

results yielded from such analyses are informative, correlations with the micro-

environments for these sites are not directly addressed or synthesized on a regional

scale. Specialized ecosystems are referenced in several reports yet these areas lack

14

definition in relation to the sites and associated prehistoric behavioral patterns. In

some cases the economic utility of the environment is addressed but these data are not

directly related back to site-specific interpretations. Most investigators refer to the

general physiographic characteristics and then shift focus to describing the archaeology.

While the archaeology is of primary importance and provides evidence of prehistoric

behavior it must not be viewed in isolation. Details concerning the specific ecological

setting where this evidence was found are equally important. The dynamic relationship

that exists between the two must be preserved. As such, the inter-relationships

between the artifacts and the immediate environmental setting should be evaluated and

interpreted accordingly. The current research is designed to reveal patterns in these

inter-relationships. In addition, relationships between site locations are evaluated with

respect to the multi-facetted context in which these sites are found, in order to consider

change through time, site density, occupation frequency and explain site location.

Objectives

All of the research efforts conducted in this region have provided important

contributions and valuable insight regarding prehistoric populations that inhabited

north-central Texas. However, when sites are viewed in isolation, regional settlement

pattern behavior is difficult to discern. Given the dynamic relationship of the parts

paired with the bulk of archaeological data, formulating a regional model is a formidable

task. Fortunately, geographic information systems (GIS) technology provides

organizational mechanisms to manage the wealth of data and create links between the

various components of a complex ecosystem and archaeological sites. Rather than

15

accepting the previous generalizations, concerning settlement pattern behavior in north-

central Texas, based on a few excavated sites or surveys, the objectives of this

research are to:

1. Develop research methods and integrative techniques that enhance evaluation, analysis and interpretation of regional settlement patterns.

2. Conduct analysis of a large collection of prehistoric surface sites using

excavated sites, as a means of temporal control, in order to evaluate settlement models formulated for this region.

3. Transform environmental data layers into interpretive maps for spatial and

statistical analysis.

4. Employ GIS to define site catchments, reconstruct the economic resource space around sites, extract descriptive information and attach meaning to these attributes for interpretive analysis.

5. Test a series of hypotheses with spatial and statistical techniques to address

unanswered questions and evaluate settlement pattern theories proposed by previous investigations.

6. Provide more specificity for the interpretation of regional settlement patterns

by utilizing GIS techniques to manage large data sets, identify the particular variables that may have attracted prehistoric populations to sites and investigate adaptive changes through time.

7. Develop a methodology for future investigations and a model to explain

regional prehistoric settlement pattern behavior, in north-central Texas, that can be utilized beyond the study area, in other regions, to improve cultural resource management practices.

8. Create environmental archaeology maps to facilitate future cultural resource

management, planning, mitigation, preservation, excavation, and analysis.

Primarily, this research is conducted to shed light on prehistoric socio-economic

behavior by constructing a regional environmental archaeology database and

performing SCA to evaluate previous investigations in an interpretable environmental

setting. Focus is upon decision-making processes, related to site selection, as reflected

16

by the variability and interpretive value of soils, vegetation, faunal remains and

artifacts. To meet the objectives of this research, GIS software is utilized to place

archaeological sites in their spatial context, establish economic territories, identify

associated environmental variables, infuse the environment with interpretive meaning,

correlate artifacts and statistically evaluate a series of hypotheses. These data are

interpreted and results are synthesized in a regional model of prehistoric settlement

patterns in north-central Texas.

17

CHAPTER 2

METHODOLOGICAL FOUNDATIONS

Essentially, settlement patterns are correlates of prehistory (the archaeological

record) and geography (spatial information). Prehistory and geography are closely

linked with respect to the fact that both examine the distribution of human phenomena,

use of natural resources and inter-relations of humans with their environment (Renfrew,

1969: 74). Site catchment analysis (SCA) provides a means to evaluate these

relationships and cultural ecology, a means to interpret the associated behaviors.

Cultural Ecology

Interpretation of prehistoric behaviors is interwoven with cultural ecology. As a

science, ecology is concerned with investigating the reciprocal relationships between

living populations and the natural and social environment in which they exist (Geier,

1974: 47). Cultural ecology is “adaptation to environment” (Steward, 1955: 30). The

differentiating factor between ecology and cultural ecology is related to the added

dimension of cultural behavior which forms a dynamic part of the ecosystem. According

to Geier, this approach is more than studying relationships between man and nature,

It is concerned with the patterned interrelationships of individuals within human populations, the manner in which a population interacts with each other and with different living species; and it is concerned with the effect that change in the physical environment has on influencing and restructuring the relations of individuals and the relations between human and non-human populations in an eco-community (Geier, 1974: 47).

Since the time of Darwin, the environment has been conceptualized as the “total web of

life wherein all plant and animal species interact with one another and with physical

features in a particular unit of territory” (Steward, 1955: 30). Understanding the

18

mechanisms of cultural development begins with variation. This variation is what

defines the unique character of individual cultures. More important then the variation

itself is the source of the variation and that it can be correlated with psychological

preferences or ecological adaptations. It is this variability that explains behavioral

change and this is precisely what archaeologists seek to do.

We will never learn what causes variability in the archaeological record by simply providing a behavioral “interpretation” of our sites or if we are fortunate to have chronological data, “by reconstructing culture histories.” We must strive to learn what factors have conditioned cultural variability in the recent as well as in the distant past. This goal requires that we embrace research strategies that permit us for purposes of learning to use our prior knowledge analytically rather than simply apply it to the archaeological record in a accommodative piecemeal fashion (Binford, 2002: 472).

SCA provides a means to evaluate the ecological factors that have influenced cultural

behavior. According to Jochim (1979: 84), “until we understand the mechanisms

involved, we are limited to description and analogy.” Geier aptly points out that the

“description of a behavior pattern does not tell us why it exists only that it exists”

(Geier, 1974: 46). Cultural history is not fully interpretable without the incorporation of

the ecological framework into research strategies,

For archaeological interpretation to proceed past mere description and notation of observed patterns and regularities in material debris, the frame of reference must be extended beyond the artifact to site context, or beyond the site to the social system, eco-community, and eco-system in which it occurred. To describe a behavioral event is only one step; to understand it is the real goal (Geier, 1974: 47).

An ecological “frame of reference” (Geier, 1974; Binford, 2002) can help understand a

portion of the mechanisms involved that influence variable prehistoric behavioral

choices in relation to site selection. Developing an ecological understanding involves

19

“exploration of the resources used in relation to those available, the constraints on the

organization of procurement, the spatio-temporal variability of the environment, and

the environmental effects of exploitation” (Jochim, 1979: 89). By employing an

“ecological frame of reference”, human culture is seen as an integral part of the

environment or ecosystem, influencing and being influenced by, the patterned

interaction of its parts (Geier, 1974: 47).

Ecosystem complexity has prompted trends of ecological research in archaeology

that require multi-scalar research. Jochim (1979) cites several examples of

archaeological attempts to reconstruct prehistoric economies that correspond with this

trend. According to Jochim (1979), it is beneficial to limit the number of interrelated

variables in order to examine their interaction and reaction to change. These allegorical

windows provide brief glimpses at isolated parts of the archaeological record. In some

respects, it is necessary to break the system down into smaller components in order to

understand the parts that create the whole. However, although, the information

derived from specialized studies is of great value, it must be viewed as part of the

larger system. Focus on an isolated part causes distortion of the larger picture whereas

focusing solely on the larger picture obscures important descriptive and interpretive

details. As such, a healthy balance between these two perceptual viewpoints is

desirable. For settlement pattern studies to be potent, empirical details along with an

understanding of their cumulative effects must be interpreted, translated into

subsystems and evaluated in relation to other subsystems that form patterns within the

larger system.

20

A common problem inherent in utilizing a theory of cultural ecology relates to the

concept of “environmental determinism.” According to Steward (1938), incorporating

ecology is not environmental determinism or economic determinism. It is a

“methodological approach” to gain a deeper understanding of behavior patterns

required to survive in a given environment. Steward (1938: 261) views it as an

“equation of culture process involving the interaction and mutual adaptation of both

historically and environmentally determined behavior” rather than “a claim a priori that

ecology predetermines cultural behavior.” Even in the sciences the realization that

“physical scientific laws are not deterministic but only statistical approximations of very

high probability based on finite populations” has been accepted (Haggett, 1966: 26).

This is a key point in support of the current research. Concern is with probability

patterns that emerge from the landscape, in association with the archaeological record,

as a basis for the interpretation of regional settlement patterns. The effect of the

environmental setting on the range of possibilities available for the manifestation of

human behavior is what is appealing about an ecological approach. Even though

different economies, each composed of a system of activities that include the utilization

of different resources, may be employed within a specific environment there are still

parameters that set limitations. Any adaptation may vary only within the limits of the

character of the natural environment or the populations cannot survive. Establishment

of these limits is an important part of interpreting prehistoric behaviors and spatial

patterning. Combining cultural ecology precepts and geographical theory is the basis for

establishing these parameters and understanding the associated dynamic relationships.

21

Geographical Theory

In order to examine the spatial patterning of past human activities archaeology has

borrowed a number of analytical techniques from geography and adapted them

successfully. Locational theory, “a product of economics, ecology, systems theory, and

regional analysis” (Haggett, 1966: 26), has been applied to prehistory with little change

from its use in human geography. The text, Locational Analysis in Human Geography

served as a “major contribution to prehistoric archaeological theory” (Renfrew, 1969:

74). Central place theory first proposed by Von Thunen, in the early 1800s, developed

into a method of geographical analysis that became central to locational theory.

Von Thunen hypothesized that an isolated population center, in a uniform

environment, with an ideal distribution of production would create distinct land-use

patterns that form concentric circles radiating from the center outward. Essentially, the

further from the base site resources are, the greater their economic cost. Eventually

there is a point where cost surpasses the return. This threshold defines an economic

boundary that forms the exploitation territory of a site. A similar view developed by

Christaller (1966) incorporates site hierarchies and their related economic spheres. The

range of goods and resources available to a particular settlement are the foundation of

the hierarchies. These theories of settlement hierarchy, referred to as “central place

theory”, proposed ways in which a settlement system could be spatially organized to

perform certain types of work most efficiently (Christaller, 1966).

The location-allocation model developed by Bell and Church (1980) was formulated

to evaluate archaeological settlement patterns by assigning “relative weights to

22

strategic criteria, resource constraints, principles of economic efficiency, political

control, and transportation interconnection” (Butzer, 1982: 222). The assumption is

that settlement patterns are the “result of political, economic, and ecological forces

working themselves out on the landscape” (Butzer, 1982: 222). Jochim (1976)

developed a gravity model to analyze the interactions between human population and

several preferred resources. It operates on the premise that site locations are situated

near high density resources associated with decreased mobility (Jochim, 1976). Butzer's

(1982: 223) ecologically based resource concentration models offer effective

explanatory mechanisms that incorporate “differential plant productivity and animal

biomass of different biomes.” Such models “imply higher group density in preferred

biomes, creating stepped population gradients between marginal and optimal

environments, with major discontinuities at or near the boundaries” (Butzer, 1982:

223). These resource concentration models are viable mechanisms for forming

hypotheses and developing theories about the dynamics of hunter-gatherer behavior.

Numerous applications, of geographical theory, to archaeological problems have

been investigated. These geographically based models provide multiple perspectives

and means to observe behavior in a spatial context. Human organization across the

landscape ultimately reflects a strategic approach to survival. Geography provides

effective tools and perspectives for archaeologists to evaluate prehistoric settlement

patterns. Vita-Finzi, a geographer, and Higgs, a prehistorian, realized the benefits of

collaboration between the two disciplines when they employed a method of locational

analysis, which resulted in the development of the concept of SCA.

23

Site Catchment Analysis

Geographical theory and cultural ecology collectively form the foundation for SCA.

Central place theory forms the primary theoretical geographical foundation. Cultural

ecology includes fundamental procedures that are integral to SCA. An analysis of the

interrelationship of material culture and environment is necessary. Behavior patterns

involved in the exploitation of a particular area by means of a particular technology

must be analyzed. Environmental variables include: geology, topography, soils,

hydrology, vegetation, climate and fauna. Some subsistence patterns “impose very

narrow limits while others allow considerable latitude” (Jochim, 1979: 40). In addition,

the extent to which behavior patterns involved in exploiting a particular environment,

affects other aspects of culture, must be determined. This aspect is much more

abstract and can only be inferred based on the other two inter-relationships. Human

behavior in its various forms of interaction requires study areas of different sizes.

Three categories formulated during previous investigations include: “those focusing on

the site and its catchment, the larger sustaining region for a group and broader areas

containing several groups or larger societal units” (Jochim, 1979: 88). The current

research employs a multi-scalar approach. SCA is performed on 150 sites. These sites

are spatially and temporally analyzed to formulate explanations for regional settlement

patterns.

SCA is a method that has been utilized by archaeologists to correlate prehistoric

cultural remains with reconstructed environments (Vita-Finzi and Higgs, 1970; Schiffer,

1976; Butzer, 1982). Tiffany and Abbott (1982: 314) define SCA as an “inventory of

24

artifactual and non-artifactual remains in relation to their sources.” Its utility derives

from the “incorporation of techniques relating site location to resource availability within

a seasonally variable territory immediately surrounding a site” (Roper, 1979: 135). It

assumes that this territory is of primary importance in structuring settlement patterns.

This type of analysis provides an exploratory and explanatory mechanism for the

organization of settlements upon the landscape.

The concept of site catchment developed in response to an expressed need for

integration between environmental and archaeological aspects of the record (Vita-Finzi

and Higgs, 1970). The term “site catchment analysis” was first used in archaeological

literature by Vita-Finzi and Higgs to describe “the study of relationships between

technology and those natural resources lying within economic range of individual sites”

(Vita-Finzi and Higgs, 1970: 5). In other words, it is a technique that has been

employed to analyze the locations of archaeological sites with respect to the economic

resources available to them. SCA was derived from the movement-minimization

concept, first introduced by central place theory (von Thunen, 1966), that land use is

determined by transportation costs. SCA is typically performed by “delineating an

arbitrary economic territory surrounding a site and evaluating the resource potential

within that area” (Tiffany and Abbott, 1982: 314). Simplistically, the exploitation

territory or catchment of a site can be approximated as a circular area centered on the

site in question. The content of the catchment is interpreted by “comparative

knowledge of site territories, resource distribution, site contents and settlement system

structure” (Higgs and Vita-Finzi, 1972: 27).

25

Depending on the location, climate and time period, the environment offers very

different possibilities. Typically, people are willing to travel only so far to utilize these

resources (Roper, 1979). Sites frequently occupy positions that are not typical of their

general zonal setting.

Man sought the optimum and hence the atypical situation to meet his needs. Even if sites should prove after study to occur in an environmentally typical situation, it cannot be safely assumed at the outset that their presence is due to the factors that determine the environmental type. One means of ensuring that the environmental attributes investigated relate primarily to the site, its occupation and its occupants is by the intensive study of the small area immediately adjacent to it (Vita-Finzi and Higgs, 1970: 62).

Although this view provides information about the immediate vicinity of the site it is

important to note that exploitation territories of prehistoric sites extend beyond the

immediate vicinity especially in relation to a mobile economy.

Limitations are inherent in zonal approaches in terms of generalization but also in

catchments due to the increased mobility of hunter-gatherer populations. The

originators of the concept recognized these limitations. A human group that exploits

more than one site territory during a year may vary the intensity and focus of its

activities at certain sites. As such, it is the annual territory, rather than an individual

site territory, that accurately reflects the economic system (Jarman, 1972; Yellen,

1977). When micro-environmental settings are investigated it is important to keep in

mind the actual sphere of influence, of the particular population represented at the site,

which likely extends beyond the area (Vita-Finzi and Higgs, 1970). Therefore, a multi-

scalar approach that involves within-site and between-site investigations is necessary.

26

It is first essential to identify the territory in the proximity of a site, where primarily

subsistence needs would be met, and then expand outward. Variable techniques have

been employed to define catchment areas. Distance is one of the core variables in that

it is the most fundamental property of spatial data. Estimates of where the boundaries

of these territories should be placed have been derived from ethnography and differ

depending on the cultural economic base and landscape character. Studies of modern

hunting and gathering economies have indicated that the territory exploited from a site

tends to be within certain well-defined limits. “Range of possible behavior is both in

theory and practice limited by the capacity of a human population to exploit an area

effectively, and is further reduced once the concept of optimum is introduced” (Jarman

et al., 1970: 62). Other things being equal the further the resources are from the site

the less likely that they will be utilized. Depending on the technology available at the

time, beyond a certain distance, the area becomes uneconomical and is unlikely to be

exploited (Chisolm, 1968). Ethnographic studies indicate that the base site will be

moved, once the economic threshold has been met, to increase efficient utilization of

resources (Lee, 1968). It is important to keep in mind that these prehistoric

populations traveled by foot. “For optimal accuracy, distance should be qualified in

terms of local topography; the operative factor is the time and effort involved in

traveling the distance, rather than the absolute distance itself” (Jarman, et al., 1970:

63). Recognition of the inter-relatedness of these factors paired with the minimization

of work as a motivating factor in human behavior is the basis for constructing site

catchment parameters.

27

Studies of contemporary communities verify the importance of time, effort and

distance in defining site territories (Lee, 1968). Chisholm (1968) and Vita-Finzi and

Higgs (1970) both create an area that can be reached within 2 hours walking time, that

takes into consideration the effects of topography and accessibility, for hunting and

gathering economies. This revision of Von Thunen’s model uses measurements of

travel time rather than linear distance. Energy expenditure is considered to be more

crucial than distance in determining the limits of exploitation territories. Two-hour radii

are established on the grounds that daylight hours remaining after traveling to the

outer threshold and back would be sufficient for food preparation. Resulting territories

were compared with ethnographic case studies and found to fit fairly well with the size

of the territories exploited from home bases respectively by collecting communities.

A combination of studies by Chisholm and Lee produce similar conclusions

concerning the relationship of production to distance. These conclusions are that

“available resources become uneconomic beyond a distance of 10 km” (Chisholm, 1968:

7). Site exploitation territory, for the modern hunter-gatherer !Kung Bushmen culture,

is within a radius of about 10 km. This 10 km radius is “consistent in many parts of the

world despite technological and environmental differences” (Vita-Finzi, 1978: 26). In

Hastorf’s analysis, the calculation of site catchment is based on the maximum radius

beyond which a particular strategy is not effectively pursued. This study uses a 4km

radius for everything except nut gathering and large game hunting. “Large game

hunting and nut gathering encompassed a 10 km radius” (Hastorf, 1980: 90). Most

studies have used either circular territories of fixed radii or time contours. Particularly

28

in the case of flat or relatively uniform environments territories tend to be circular. A

variety of site catchment models have modified the parameters of the economic

boundaries, in different ways, based on the character of their region.

Site Catchment Analysis Case Studies

SCA studies have been conducted by various archaeologists over the past thirty

years to analyze the environmental context of sites, determine site function and explain

site distribution (Vita-Finzi and Higgs, 1970; Webley, 1972; Schiffer, 1976; Zarkys,

1976; Ferring et al., 1978; Flannery, 1976; Roper, 1979; Tiffany and Abott, 1982;

Butzer, 1982; Bernick, 1983). SCA has also been utilized as a means to determine the

“feasibility of various forms of economy, modeling settlement patterns and the study of

demographic processes” (Roper, 1979: 135). Inferences regarding site function have

proven useful in comparing the locations of sites. Contemporary site territories may

have resources or properties in common, or may be economically complementary (Vita-

Finzi and Higgs, 1970). Essentially, SCA focuses on the ecological, economic and

procreative aspects of human behavior that extend beyond the archaeological remains

(Hunt, 1983).

Vita-Finzi and Higgs (1970) utilized SCA techniques to establish the economic

boundaries of sites in Palestine and systematically study the resources in this area.

They classified the land within these boundaries into categories of prehistoric land use

in order to make predictions concerning resource use. Land use classifications that are

established for the current research include edibility, medicinal value, supplemental and

overall economic categories. According to these authors, some site locations are not

29

chosen primarily because of their land potential. When site locations do not correlate

with areas that have a high economic potential other factors are considered to look for

explanations for site selection. A similar approach is utilized in the current research.

Sites that lack explanation in isolation can acquire meaning when compared with other

sites (Vita-Finzi and Higgs, 1970). In addition, the character of the landscape combined

with the climatic conditions that were prevalent at the time provides clues. By

conducting intra-site comparison it has been shown that two sites situated 30 miles

apart were complementary to each other in that they provided suitable seasonal

settings for a hunter gatherer economy (Vita-Finzi and Higgs, 1970). Based on the

character of prehistoric site distribution in north-central Texas it is expected that

patterns similar to these occur in this region. SCA has also revealed that two different

cultures, as defined by their artifacts, may represent two different economies rather

than two different groups (Vita-Finzi and Higgs, 1970). Sites that are a part of the

current research are likely to contain similar artifacts. Therefore it is expected that the

economies offered by particular locations on the landscape are the primary source of

differentiation between site locations.

Environmental diversity indices for site catchments have been utilized in previous

investigations to measure the potential of an area for the presence of archaeological

sites and to determine the extent and intensity of human use (Tiffany and Abbott,

1982). The basic assumption is that site presence is correlated with increased

environmental diversity. The idea of establishing an environmental diversity index is

based on the premise that,

30

the biological and cultural needs of a society are easier to fulfill in a natural environment that has the greatest environmental diversity per area of catchment for that society. The occurrence of an archaeological site on a given portion of the landscape should be a function of the diversity of the local natural environment and likewise the greater the potential diversity of a particular study area, the more extensive and complex will be its use by groups (Tiffany and Abbott, 1982: 315).

Tiffany and Abbott (1982) consider reoccupation of a site as a function of the stability

of the local environmental diversity through time. The environmental diversity index

employed by these authors involves counting the diversity of mineralogical, hydrological

and vegetational resources within the site catchment territory. The current research

utilizes a diversity index to test for this however it is derived by different means.

Hunt conducts a case study in Northeastern America by utilizing a SCA methodology

that includes Geographic Information Systems (GIS) based analysis (Hunt, 1982). Hunt

is concerned with the landscape that surrounds the archaeological site and the concept

of site catchment as it applies to quantifying the physiographic space within and around

a site (Hunt, 1982). Hunt demonstrates the ability to characterize soil matrix around

Late Woodland village sites in a multitude of ways and to link soil data to other

attributes (Hunt, 1982). Webley’s (1972) analyses of Tell Gezer and several other sites

in Palestine are based entirely on analysis of soils and their potential productivity

(Roper, 1979: 126). While such studies are limited to description of how sites are

located relative to soils or vegetation, soils analysis offers valuable information to

cultural history, including comparisons among sites of their potential for certain

economic activities. According to Roper (1979), focusing on soils or vegetation in

isolation may not explain why a site is located where it is or how it functioned in a

31

settlement system. More complete modeling of settlement location and the settlement

system requires the use of a wider variety of resource types (Roper, 1979). While the

current research is highly dependent on soils, to economically interpret site catchments,

other factors are taken into account to provide explanatory mechanisms. Each

component is a piece of the puzzle that fits together to form a more comprehensive

view of prehistoric behavior.

In a study about subsistence strategies of prehistoric populations that inhabited the

Mimbres River Valley in New Mexico, SCA is the principal technique used to collect and

systemize information (Hastorf, 1980). In this study conducted by Hastorf (1980) six

food gathering strategies are described, and by SCA, potential yields for each strategy,

in each time period, are determined. The economic model provides the framework

within which the strategies and their potential yields are organized and graphically

depicted. According to the model, “in a subsistence economy, population size

determines both caloric requirements and marginal costs needed to meet those

requirements” (Hastorf, 1980:80). The assumption is that there is a definable area for

each food-procuring strategy associated with each settlement (Hastorf, 1980). SCA

translates environmental information into potentiality for particular strategies. SCA is

used to determine a land-use area for each strategy for each time period. Once this

area is defined and the caloric productivity per unit is calculated, the potential yield for

each strategy is obtained (Hastorf, 1980). Given the character of the prehistoric sites in

north-central Texas this type of approach is not directly applicable for the current

research.

32

Presence of the unique floral and faunal material excavated at a site on Vancouver

Island, in British Columbia, prompted a site catchment analysis (Bernick, 1982).

Instead of starting with the site territory and reconstructing the economic patterns, an

analysis conducted by Bernick (1982) begins with the set of excavated cultural remains

that imply particular subsistence strategies. The exploitation territory is delineated

based on these data. Her objective is to define the spatial dimensions of economic

activities and to assess the applicability of the site catchment model to analogous sites

on the Northwest Coast (Bernick, 1982). She factors in a defined travel time required

to procure specific resources and the topographical location of sites. Confirmation of

hypotheses is interpreted as an indication that the site catchment model is effective for

evaluation of economic behavioral patterns of Northwest Coastal prehistoric

communities (Bernick, 1982). This case study is an excellent solution for site specific

investigations. However, on a regional scale, currently the data are lacking to

successfully carry out a methodology such as this one.

Statistical techniques are an effective mechanism for objectively evaluating site

catchments in relation to their immediate resource potentials. Previous SCA studies

have utilized statistical techniques to accomplish this goal. Multivariate statistical

techniques such as “factor analysis, multidimensional scaling and cluster analysis” are

utilized by Roper for description and comparison of site territories with their resource

potential (Roper, 1979: 127). Zarkys (1976) analysis of site catchments at Ocos in

Guatemala used “percentage point differences, chi-squared tests, and binomial tests for

evaluating which environmental zones were represented in higher proportions

33

immediately surrounding a site than they were in the total study area” (Roper,

1979:128). Kvamme and others followed a similar methodology comparing site

selection to overall regional characteristics incorporating a GIS to facilitate analysis,

A raster GIS, together with quantitative analysis software by the SAS institute Inc. [1985], provides an ideal tool for spatial analysis because a variety of archaeological and biophysical data can be co-registered to vast numbers of locations cross broad regions, allowing rapid investigation of relationships between archaeological features and any number of locations across broad regions, allowing rapid investigation of relationships between archaeological features and any number of environmental parameters (Kvamme, 1989: 168).

A variety of statistical approaches have been employed to condense vast amounts of

data into interpretable statistics. However, it is important to note sample size and

archaeological recovery bias. Particularly, in the current research, a strong bias is

already present in the amount of contract work that has been conducted along the Elm

Fork of the Trinity due to reservoir construction. Considerably less work has been

conducted in Collin County and even less in Wise County. As a result, the current

statistical results are heavily influenced by recovery bias.

SCA has proven to be a very useful methodological technique that has provided

answers to a plethora of questions to meet the objectives of a variety of research

efforts. Ultimately, SCA provides a means to develop a framework for investigating the

distribution of resources in relation to site location. The rationale for SCA operates on

the assumption that,

human beings are refuging animals, rhythmically dispersing from and returning to a central place (Hamilton and Watt, 1970: 263), differentially using a seasonally and spatially variable landscape in a manner that generally is conservative of energy, but conservative relative to a relative scale of values placed on needs and wants (Roper, 1979:122).

34

Given that such behavior takes place in particular locations it is logical to extract and

evaluate environmental attributes from catchments of archaeological sites to shed light

on prehistoric behavior. The theoretical foundation of the SCA model is a sound and

viable basis for the current research. This method is a useful way to integrate the

archaeological investigations conducted in the region and a means to systematically

study sites in relation to the dynamic character of the ecosystem.

Although the current research employs SCA techniques that are similar to those that

have been utilized in previous investigations it does not follow any of these research

methodologies in particular. Given the data limitations and character of the region, the

current research design incorporates a variety of techniques suggested by several

studies, modifies them and adapts them to meet the needs of the current research.

This methodology is designed to provide answers to the unanswered questions related

to prior research conducted in the region and formulate a model of prehistoric

settlement pattern behavior.

Geographically, factors such as topography, geology, soils, vegetation, water

sources, fauna and land suitability influence past cultural behavior. Identifying the

specific cultural and non-cultural characteristics that factored into site selection with

respect to site catchment variability is pertinent to understanding the dynamic socio-

economic prehistoric behavioral patterns in north-central Texas. In order to gain a

deeper understanding of these patterns it is appropriate to begin with an evaluation of

the natural setting (environment) followed by the synopsis of the defining

characteristics of the various populations (cultures) that inhabited the region.

35

CHAPTER 3

RESEARCH SETTING

North-central Texas is located in an advantageous position within the “bioclimatic

transition between the forested Gulf Coastal Plain and the rolling Southern Plains”

(Ferring, 1997:19). Increased “biological diversity” is the result of numerous factors,

including the region's geologic, soil and climatic variation combined with its eco-tonal

location between the “deciduous forests of East Texas and the grasslands of the

Southern Plains” (Diggs et al., 1999: 20). The combination of these factors provides a

highly productive blending ground for plants and animals from the east and west. In

addition, the increased diversity of this region offers a wide range of economic

opportunities for prehistoric populations.

Geology of North-Central Texas Geological structures in north-central Texas were shaped by uplift and formation of

an extensive mountain system that includes the “Appalachians, Wichitas and Ouachitas”

(Diggs et al., 1999: 21). Through time, shallow seas repeatedly advanced and

retreated over most of Texas depositing a number of different layers of Cretaceous

sediments (Diggs et al., 1999). As a result of these depositional processes and

subsequent erosion, surface rocks in north-central Texas are predominantly Cretaceous

in origin. Numerous layers of limestone, marl, shale, and sandstone were deposited

over the area. Since climatic variation in north-central Texas is minor, differences in

landforms, soils and vegetation are attributed to bedrock lithology (figure 2) (Ferring

and Yates, 1998: 5). Lynott (1981) describes the geology as, “dominated by a series of

36

eastward dipping sedimentary marine deposits.” Pennsylvanian deposits are exposed in

the western parts of the region, while Cretaceous clays and sands dominate the eastern

prairies. These eastward dipping strata (figure 2) consist of Antlers Sandstone, Denton

Clay, Woodbine Sandstone, Eagle Ford Shale, Austin Chalk and Ozan Marl interspersed

by resistant outcrops known as the Goodland, Kiamichi, Pawpaw, Weno, and Grayson

formations. The entire Upper Trinity drainage basin has developed over these

Cretaceous and Pennsylvanian sedimentary rocks (Ferring, 2001; Diggs et al., 1999).

North of the West Fork of the Trinity the “Antlers Formation is primarily composed of

fine-grained sandstone and shale” (Ferring and Yates, 1997: 19). South of the West

Fork the “Twin Mountains, Glen Rose and Paluxy Formations have a more diverse

lithology” (Ferring and Yates, 1997: 19). The upper part of the West Fork drainage,

northwest of Fort Worth, formed on top of Pennsylvanian limestone, shale and

sandstone. All other portions of the Upper Trinity drainage basin developed over

Cretaceous rocks (Ferring, 2001). Geology is culturally significant in that it provides

resources for tool manufacture and settings for sites. In addition, the characteristics of

the geology influence the character and range of potential soil types that can develop.

Geology, combined with the climate, organisms, and topography ultimately determine

the range of possible vegetation that can be produced in a given area. In north-

central Texas limestone settings have predominantly produced prairie vegetation and

sandstone settings have provided the foundation for forest vegetation. The variation

within these geological formations is illustrated (figure 2) from a bird’s eye view.

37

Figure 2. Geology of north-central Texas (based on USGS; Ferring and Yates, 1997).

38

Topography

The topographic expression of geological formations in north-central Texas

originated from tectonic activity, erosion and variable depositional processes. Elevation

in north-central Texas generally decreases from west to east. Topography is nearly

level to very hilly. Wise County is transected by the Fort Worth Basin which combined

with the uplifted resistant bedrock creates a highly variable landscape with steep

slopes. The elevation reflects this variability ranging from 1,286 feet in the southwest

to 649 feet in the eastern portion of the County. Denton County, although also

transected by the Fort Worth Basin in the southern portion of the county, is less

variable with very few resistant outcrops. This is reflected by elevations that range

from 900 to 500 feet. Elevations, in Collin County, reflect the less resistant limestones,

shales and marls, range from 900 to 400 feet, gradually decreasing eastward. Collin

County is situated in an area where the Fort Worth Basin and East Texas Basin merge.

This topographic setting reflects relatively lower elevations that explain the increased

number of sites associated with bottomland settings along the East Fork of the Trinity.

Topography in north-central Texas combined with the character of the geology

ultimately influences and is influenced by the character of the drainage networks.

Drainage Systems

Drainage systems have actively sculpted the landscape forming as a result of the

topography and resistivity of the geology. Northern Texas is drained by a network of

rivers that includes the Brazos, Canadian, Colorado, Red River, Sabine, Sulphur and

Trinity (figure 3). One of the most dominant features transecting the landscape of

39

F

igure 3. Hydrological networks of north-central Texas (based on TNRIS data).

40

north-central Texas is the Trinity River. From its beginning, northwest of Fort Worth in

Archer County, the Trinity flows to the southeast approximately 692 River miles to the

Gulf Coast with a change in elevation of 1,250 feet (Nixon and Willett, 1974). The

Upper Trinity drainage basin is surrounded by three major drainage basins: the Red

River to the north, the Brazos River to the southwest, and Sabine River to the east.

The Trinity has been a focal point of cultural activity for thousands of years.

The West Fork of the Trinity flows in a southeasterly direction through Jack, Wise,

Tarrant and Dallas Counties until it forms confluence with the Elm Fork of the Trinity.

The West Fork is a consequent stream of the Upper Trinity drainage basin that has

incised into the resistant Woodbine sandstone and Austin chalk (Ferring, 1998).

Although archaeological sites are rarely reported, along the West Fork, within the

confines of Wise County, there is considerable undocumented evidence to suggest that

prehistoric populations aggregated along this major tributary. Denton Creek is a major

tributary in the Upper Trinity Basin that forms a part of the Trinity “fork” system. It is

referenced due to its archaeological significance and proximity to the West Fork. The

headwaters of Denton Creek begin in Montague County and flow to the southeast

through the northeast corner of Wise County and the southwest corner of Denton

County where it veers east to form a confluence with the Elm Fork. For the current

research sites near Denton Creek will be included in the West Fork study area

The headwaters of the Elm Fork of the Trinity are in the west central section of

Montague County near the Cooke County line. From this point it flows in an easterly

direction to central Cooke County where the principal drainage curves to the south and

41

then southeast. This general course is maintained to the stream's confluence with the

West Fork of the Trinity near Dallas. The vast majority of sites in this region are

associated with the Elm Fork of the Trinity.

The headwaters of the East Fork of the Trinity originate in Grayson County flowing

south through Collin, Rockwall, and Grayson Counties until Ellis County where it forms a

confluence with the Trinity. This major tributary is situated in a unique environment

centrally located in the Blackland Prairie where it serves as the only major intermediary

water source between the main stem of the Trinity and the Sabine River.

All of these drainages follow the tilt of the regional bedrock forming elongated

dendritic patterns. The West Fork, Denton Creek, Elm Fork and East Fork all merge into

one by the time the River system reaches Ellis County where the Trinity guides the flow

to the Gulf of Mexico. In prehistoric times drainage systems acted like highways linking

separate geographical areas together. In addition to being culturally significant, these

hydrological networks are responsible for the variable character of the landscape.

Physiographic Setting

Four major upland physiographic subdivisions (figure 4) have been recognized in

north-central Texas: the Western Cross Timbers, Grand Prairie/Fort Worth Prairie,

Eastern Cross Timbers and Blackland Prairie (Dyksterhuis, 1948; Ferring and Yates

1997). The character of the four major upland geomorphic subdivisions is influenced

mostly by the bedrock lithology and the correlative soils. These alternating strips of

prairies and oak woodlands form a landscape that is mosaic in character and has

significantly contributed to the increased diversity of north-central Texas.

42

Figure 4. Physiographic regions of Texas and north-central Texas (TNRIS).

43

The Western Cross Timbers is a physiographic subdivision that occurs in the western

part of the Upper Trinity Basin. This region, situated mainly in the Wise County portion

of the study area, is a rolling to deeply dissected area with sandy soils that corresponds

with the Antlers Formation (Ferring and Yates 1997; Ferring 2001). It is characterized

by a rolling to hilly topography with a very sandy soil cover. The Western Cross

Timbers are similar to the Eastern Cross Timbers, yet less rugged in character due to

the corresponding geology.

The Grand Prairie physiographic subdivision occupies the central part of Denton

County and the eastern portion of Wise County. Alternating beds of shales and

limestones stratigraphically situated between sandstone formations provide the

geological foundation. Typically, it is a “rolling upland prairie with small escarpments

and benches” formed by geological layers such as the Goodland formation, Duck Creek

limestone, Fort Worth limestone and Main Street limestone (Skinner, 1973: I-II). The

Fort Worth Prairie is a level to gently rolling surface, nested in the central portion of the

Grand Prairie, north of the West Fork of the Trinity. The foundation of this area is a

resistant Cretaceous limestone, composed of the Goodland Formation in the west and

Grayson Formation in the east. “Differences in lithology have primarily controlled

development of the upland prairie topography and tributary alluvial valleys” (Ferring,

2001: 27).

The Eastern Cross Timbers are situated along the central and eastern boundary of

the Upper Trinity Basin. This physiographic region follows the Woodbine sandstone

forming a narrow north-south belt of low hills that rise above the prairies on either side

44

(Ferring and Yates, 1998). Due to the fact that the Woodbine formation is not as thick

as the Antlers Sandstone, the Eastern Cross Timbers are narrower than the Western

Cross Timbers (Ferring and Yates, 1998). The woodbine formation consists of

sandstones, shales, sandy shales, and fine to coarse sands providing the foundation for

the Eastern Cross Timbers. Since the sandstones are more resistant to erosion than the

clays and shales “the topography is more hilly and rugged in character” than the

Western Cross Timbers (Prikryl, 1990: 6).

The Blackland Prairie physiographic subdivision is immediately east of the Eastern

Cross Timbers. At the eastern margin of the area the “Blackland Prairie developed on

relatively young Cretaceous layers” (Diggs et al., 1999: 21). Most of the Blackland

Prairie corresponds with Upper Cretaceous rocks, including the Eagle Ford Formation

(shale, clay, and marl), the Austin Chalk and the Ozan Marl (Ferring and Yates 1997;

Ferring 2001). This physiographic region is characterized by a fairly flat to rolling

surface developed on the Eagle Ford Shale and Austin Chalk formations (Dyksterius,

1948). These patterns reflect the dominant influence that the bedrock has over the

region.

The Western and Eastern Cross Timbers, interspersed by prairie settings, form a

diverse physiographic region that is related to the geology, topography, soils and

vegetation. These physiographic settings are described in more detail in the vegetation

section following a brief review of the soils formed in north-central Texas. The

character of the soil is highly dependent on the geology and forms the basis for the

range of vegetation available for prehistoric utilization.

45

North-Central Texas Soils

Soils in north-central Texas vary dramatically, ranging from the erosive sandy soils

that form the foundation for the Western Cross Timbers to the highly plastic clay soil of

the Blackland Prairie. The character of the soil is interdependent upon the geological

parent material, topography, the organic and biotic community and the climate.

Comparisons of the soil taxonomy within the prairie and forest environments provide a

greater understanding of the range of variation and differences in vegetation that can

potentially form in each setting. West Fork soils are composed of Western Cross

Timbers, Grand Prairie and Fort Worth Prairie soils from west to east (figure 5A). Elm

Fork soils consist of Grand Prairie, Eastern Cross Timbers and Blackland Prairie soils

from west to east (figure 5B). East Fork soils are entirely composed of Blackland Prairie

soils (figure 5C). An overview of the great groups present in the study area illustrates

the soil distribution referred to in the following discussion (figure 5).

Western Cross Timbers Soils

Soils of the Western Cross Timbers are mainly Paleustalfs. These “Alfisols are

associated with the sandstone bedrock and sandy alluvium, whereas the Mollisols and

Vertisols are associated with the calcareous bedrock and alluvium” (Ferring, 2001: 27).

With the exception of the soils in the western portion of the Western Cross Timbers that

developed on gravelly and rocky Pennsylvania strata the soils of the East and Western

Cross Timbers mainly developed on sandy Cretaceous Woodbine strata. Mollisols and

Vertisols are both typically associated with the grassy areas of the Cross Timbers

whereas the Alfisols are associated with increased production of hardwoods.

46

Figure 5. Soil distribution in north-central Texas study area.

47

Duffau-Keeter-Weatherford, Windhorst-Chaney-Selden, Truce-Cona and Bastsil-Silawa

are general soil map units associated with the Western Cross Timbers (Ressel, 1989).

These loamy and sandy, well drained soils composed of material weathered from

sandstones and shales, formed upon a dominantly gently sloping surface on upland and

terrace savannahs (Ressel, 1989). Great groups associated with these units are

Paleustalfs and Haplustalfs (Ressel, 1989). A mixture of these older more developed

and younger minimally developed Alfisols maintain a fair amount of moisture with dry

intervals. Climax vegetation includes oaks, with an under story of grasses, peanuts,

grapes, tree fruits and nuts (Ressel, 1989). Woody and herbaceous vegetation provide

food and cover for deer, turkey, squirrel, dove, and quail (Ressel, 1989).

Grand Prairie and Fort Worth Prairie Soils

Soils of the Grand Prairie and Fort Worth Prairie have similar characteristics to the

Blackland Prairie but the shrink-swell properties of the clay are less intense. Differences

between the prairie environments can be attributed to the bedrock lithology. Mollisols

are primarily found on the Grand Prairie and Fort Worth Prairie on various limestone

layers (Diggs et al., 1999). “Mollisols are less problematic than Vertisols in terms of

shrink-swell (Diggs et al., 1999: 25) and are considered among the world’s most

productive soil due to their high native fertility. Soils on the Fort Worth Prairie vary

according to bedrock parent material. Most upland soils are clayey and calcareous

Chromusterts, Calciustolls or Haplustolls.

General soil units associated with the Grand Prairie include the Sanger-Purves-

Somervell and Venus-Aledo-Somervell units (Ressel, 1989). These units are clayey and

48

loamy, well-drained soils that formed in material weathered from limestones and marls

on upland prairies (Ressel, 1989). Related great groups include Chromusterts,

Calciustolls, and Haplustolls that indicate a mix of Mollisols and Vertisols in moist and

dry climates (Ressel, 1989). The presence of Mollisols and Vertisols indicate a prairie-

dominated environment. Yet cobble and limestone fragments contribute to a lithic

component that creates a habitat that is unsuitable for most wildlife (Ressel, 1989).

Aledo-Somervell soils are well-drained loamy soils formed in material weathered

from limestone in the Uplands (Ressel, 1989). These soils are not very common and

are focused in two particular locations near Denton Creek and Clear Creek. The Aledo

component is made of Lithic Haplustolls while the Somervell is Typic Calciustolls

(Ressel, 1989). Both are composed of Mollisols in fairly moist to periodic dry regimes

but the former has a stony character with minimum horizon development whereas the

latter is more calcareous indicating increased development. According to the soils

survey, 90% of the vegetation consists of varieties of grasses and the remaining portion

consists of various trees and shrubs (Ressel, 1989).

Sanger-Somervell, Navo-Wilson, Slidell-Sanger and Ponder-Lindale are fairly well-

drained clayey and loamy soils that formed in material weathered from calcareous clay

and marl in the upland prairies (Ressel, 1989). Chromusterts, Calciustolls, Paleustalfs,

Ochraqualfs, Pellusterts and Halustalfs indicate a diverse mixture of Vertisols, Mollisols

and Alfisols revealing the gradual shift from the Grand Prairie to the Eastern Cross

Timbers (Ressel, 1989). Vegetation associated with these soil units is dominated by

grasses and forbs (Ressel, 1989).

49

Eastern Cross Timbers Soils

Soils of the Eastern Cross Timbers are primarily Alfisols that developed from the

sandy woodbine formation. Alfisols are generally productive soils that promote the

growth of hardwood forest vegetation and crop yields. Principal surface soil

associations are summarized by the Birome-Gasil-Callisburg unit that contains well-

drained loamy soils formed in sandstone and clay parent material (Ford and Pauls,

1980). Great groups associated with this soil unit are Paleustalfs indicating a fairly

moist climatic regime (Ford and Pauls, 1980). Paleustalfs are the most dominant great

group in the Eastern Cross Timbers with minor variations related to the clay parent

material. On the sandy Alfisols, the woody plants--namely post oak, blackjack oak, and

hickory are dominant; while the grasses prevail on the tighter, more drought-resistant

Mollisols (Ford and Pauls, 1980). Oak is the most common tree reported in the Eastern

Cross Timbers (Ford and Pauls, 1980). River valleys crossing the region support a forest

of hackberries and pecans mixed with oaks on the alluvial soils (Ford and Pauls, 1980).

Blackland Prairie Soils

Soils in the Blackland Prairie are composed of the Houston Black-Austin group,

Houston Black-Houston, Ferris-Houston and Wilson-Burleson that consist of clayey soils

formed in marl and chalk parent material (Wheeler and Hanson, 1969). Primary

differences between these general soil units are found in the structural density of the

clay, susceptibility to erosion and slope orientation.

Three dominant soil orders identified in the Blackland Prairies include Alfisols,

Vertisols and Mollisols. Alfisols are less fertile than Mollisols or Vertisols and are found

50

mainly on the Eastern and Northern margins of the Blackland Prairie. Vertisols develop

mainly on the Eagle Ford Shale or rocks of the Taylor Group and are characterized by a

high clay content that exhibits increased shrink-swell properties. “Swelling and shrinking

causes deep and wide cracks” (Diggs et al., 1999: 23). The “black waxy” soils of the

Blackland Prairie are derived from rocks of limestone origin (Diggs et al., 1999: 23).

This “incredibly sticky soil has historically been particularly problematic and virtually

impassable during wet weather” (Diggs et al., 1999: 25). Drought however is

considered to have been more of a problem, with a lack of water considered a limiting

factor for humans (Diggs et al., 1999). Impermeable black clay soils, the lack of

dependable water bearing layers, and the ephemeral nature of surface streams made

the “early Blackland Prairie a particularly inhospitable environment for humans” (Diggs

et al., 1999: 29). Vegetation provides sustenance consisting of grasslands with a

scattered mixture of hardwoods and particularly pecans. These areas are associated

with bottomland hardwoods that provide a suitable habitat for wildlife.

Riverine and Floodplain Soils

Riparian corridors provide unique micro-environments that are significantly different

from the generalized landscapes of the physiographic regions. Similarities can be

attributed to bedrock lithology and differences to the influence of concentrated water

resources combined with the variation in the alluvial parent material. Riverine settings

provide increased productivity and tend to produce hardwoods that flourish in riverine

settings. Characteristic vegetation and faunal resources found along streams and

riverbanks reflects the variability of these settings.

51

The Pulexas-Balsora-Deleon unit is situated on the floodplains of the West Fork of

the Trinity in the Western Cross Timbers (Ressel, 1989). It consists of loamy and

clayey, moderate to well-drained soils that formed in alluvium. Associated great groups

suggest that these soils are primarily recent Ustifluvents and Mollisols that exhibit

minimum horizons (Ressel, 1989). Vegetation includes an overstory of oak, elm,

hackberry, pecan and willow varieties. Small grains, peanuts, melons and sorghum are

common (Ressel, 1989). These soils have a strong potential to support a wide variety

of wildlife.

Frio-Ovan soils are well-drained, clayey soils that formed in material weathered from

alluvium in the bottomlands that correspond with stream channels transecting the

Grand Prairie and Eastern Cross Timbers (Ford and Pauls, 1980; Ressel, 1989).

Related Haplustolls and Chromusterts indicate a mix of Mollisols with minimally

developed horizons and Vertisols with a high organic content (Ford and Pauls, 1980;

Ressel, 1989). These soils are concentrated in the floodplains of Denton Creek, Hickory

Creek, Clear Creek and the Elm Fork of the Trinity. Vegetation related to these soils is

composed of approximately 85% grasses with 15% representation of various trees such

as pecan and elm (Ford and Pauls, 1980; Ressel, 1989).

The Frio-Trinity soils are loamy and clayey, moderate to well-drained soils that

formed in alluvium on the floodplains in the Grand and Blackland Prairies (Ressel, 1989;

Wheeler and Hanson, 1969). The Frio-Trinity component contributes Haplustolls and

Pelluderts soils situated around portions of Denton, Catlett and Sweetwater Creeks.

Mollisols and Vertisols associated with these great groups indicate increased

52

productivity paired with a shrink-swell component (Ressel, 1989; Wheeler and Hanson,

1969). Soil units associated with riparian environments within the Blackland Prairie

include the clayey and loamy soils of the Trinity-Frio found on the floodplains and the

clayey soils of the Houston-Black Burleson found on stream terraces (Wheeler and

Hanson, 1969). Many native pecan trees are located along the stream channels and

interspersed on the floodplains (Ressel, 1989; Wheeler and Hanson, 1969). Yields from

Pecan harvests would have provided a very desirable resource with a high-caloric value

for prehistoric populations. The presence of this soil unit is so concentrated and

economically valuable that prehistoric presence in association with these soils would

strongly suggest a direct correlation with vegetation.

Floral Resources in North-Central Texas North-central Texas is an interesting and diverse region in terms of vegetation,

which responds to a combination of soils and climate. Following the physiographic

character of the landscape, vegetation in Texas is divided into several zones known as

the Pineywoods, Gulf Prairies and Marshes, Post Oak Savannah, Cross Timbers and the

Blackland, Fort Worth and Grand Prairies. Oak-Hickory forest and tall-grass prairies

transect this region. North Texas flora includes “2223 species (46% of the species

known for Texas) and a total of 2376 taxa (species, subspecies and varieties)” (Diggs et

al., 1999: 20). Due to increased biodiversity related to the Cross Timbers and Prairies

different microhabitats within the same county contain different plant communities.

The regional study area is composed of a combination of four distinct geographic

provinces including the Blackland Prairie, Eastern Cross Timbers, Grand Prairie, and

53

Western Cross Timbers. These zones extend, southward about 150 miles and

northward to the Arkansas River, as continuous bodies of forest or savannah bounded

by open prairie (Dyksterhuis, 1948). The Eastern and Western Cross Timbers reside

east and west of the Grand and Fort Worth Prairies. Resources associated with prairies

and forests offer a vast range of opportunities. This geographic expanse provides a

unique opportunity to investigate prehistoric behavioral choices, in relation to an

environment situated within an eco-tonal region that provided a diversity of resources.

Prairies

The predominantly clay-rich soils of the prairies support tall to mid-size grasses with

clusters of oak trees in the uplands and riparian communities that follow the waterways.

Vegetation along the waterways in the Prairies consists of similar understory species to

those found in the temperate woodlands of the Cross Timbers (Dyksterhuis, 1946;

Ferring and Yates, 1997). Climax vegetation in the Blackland Prairie is composed of

mixed grass prairies, with scattered stands of cedar and oak trees situated on the

"White Rock Escarpment" near the western edge of the Austin Chalk (Ferring, 1997:

22). Kindall, an explorer, in 1845, described the Blackland Prairie as “varied by small

prairies and clumps of woodland”, and the Fort Worth Prairie as “a perfect ocean” of

grasses (Ferring and Yates, 1998: 12; Diggs et al., 1999). Dark clay soils of the

Blackland and Grand Prairies sustain short grass prairies (Skinner and Baird, 1985: 2-2).

Dominant species are bluestem, speargrass and wheatgrass. Prairie vegetation includes

tall and midgrasses such as bluestems and gramas. Buffalo grass and species of three-

awn, among others, tend to increase under grazing. Over “250 species of plants have

54

been identified for these prairies” (McCormick et al., 1975: 39-40). Trees such as bois

d’arc, oak, elm, ash, pecan and post oak trees dot the landscape along stream channels

(McCormick et al., 1975). Brush species have invaded the prairie, and weedy annual

and perennial grasses have increased in number (Ressel, 1989). The Blackland Prairie

offers numerous seed bearing trees, plants and grasses. “Large herbivores traditionally

migrated through this area in the fall" (McCormick et al., 1975: 39-40). Streams,

transecting the prairies, provided “water and riverine resources including both

nutritional and medicinal trees and plants” (McCormick et al., 1975: 39-40).

Cross Timbers

North-central Texas contains an oak dominated vegetation region that is part of a

much larger area referred to as the Cross Timbers. Pollen data reveals that the “forests

of the Cross Timbers were not established until the middle to late Holocene” (Ferring,

2001: 28). In Texas, this region is composed of two strips, the Eastern and Western

Cross Timbers, separated by tall grasses and the Grand Prairie. Contrasts in vegetation

between Prairies and Cross Timbers create natural barriers and facilitate north-south

trails,

The conspicuous change in vegetation from the Blackland Prairie to the Eastern Cross Timbers served as a landmark on early maps (e.g. Holley, 1836; Ikin, 1941; Gregg, 1844; Kendall, 1845) and was discussed in many immigrant guides and early explorer accounts (e.g. Marryat, 1843; Roemer, 1849; Parker, 1856). Kennedy (1841) wrote, “This belt of timber varies in width from five to fifty miles. Between the Trinity and the Red Rivers it is generally from five to nine miles wide……When viewed from east or west, it appears in the distance as an immense wall of woods stretching from north to south (Diggs et al., 1999: 43).

55

The Cross Timbers have also been described as a “singular strip of wooded country with

an almost impenetrable undergrowth of brier” that was “broken” and “full of deep,

almost impassable gullies” (Ferring and Yates, 1998: 12). This thick vegetation, coupled

with the heavily dissected topography, led early travelers to comment on the difficulty

of traversing the Cross Timbers (Prikryl, 1990; Diggs et al., 1999). Early accounts of

the Cross Timbers also report of oak varieties with a dense under-story of saplings,

vines and greenbrier (Diggs et al., 1999). Of the dominant trees, post oak is the most

widespread. Hickory, which requires the most moisture, is the least common. “In the

1940s, wooded overstory covered 35% of the land surface in the Western Cross

Timbers” (Dyksterhuis, 1948: 337). This consisted of “63% post oak and 29%

blackjack oak with a remaining 8% composed of species such as hackberry and cedar

elm” (Dyksterhuis, 1948: 337). The primary difference between the Cross Timber areas

is found in the rate of precipitation that increases from west to east. Therefore, it is

anticipated that the Eastern Cross Timbers would exhibit a relatively increased

productivity. However, while there is a good variety in the subsistence resources of this

vegetation region, it is an area that is prone to droughts (Story, 1985). Western and

Eastern Cross Timbers are major areas of oak-hickory, with open savannah, dense

brush, post and blackjack oaks. Eastern Cross Timbers are essentially a mix of the

Western Cross Timbers and Oak Woodlands.

Vegetation of the Eastern Cross Timbers can be regarded as a western fringe area

of the eastern woodlands. Over three hundred species of plants are native to this area

(Tharpe, 1926). Forest vegetation is dominated by post oak, blackjack oak, pecan and

56

elm. Hickory is virtually replaced in the Eastern Cross Timbers by the Pecan tree. Main

components of the understory are little bluestem ragweed and elderberry. Small

patches of prairie within the Cross Timbers are dominated by grasses such as

wheatgrass, bluestem and speargrass. “Species diversity of dominant terrestrial plants

for the upland post oak forest and the streamside forest is low” (Skinner and Baird,

1985: 2-3). Cross Timbers areas contain sandy upland soils that support dense groves

of post oaks and greenbrier with an understory of big and little bluestem, switchgrass,

lovegrass, sumac, plum, many varieties of forbs, legumes and other shrubby vegetation

(Ferring, 1990; Story, 1985). In addition, these sandy areas often produced peanuts.

Post oak, blackjack oak and hackberry are also particularly abundant in wooded uplands

(Story, 1985; Prikryl, 1990). On the floodplains in the Cross Timbers, “tree cover

consists of elms, pecans, oaks, cottonwoods, and willows with an understory vegetation

that includes grasses, a variety of forbs and various fruit-bearing plants such as plums

and berries as supplements to a rich variety of nut-producing trees” (Ferring and Yates,

1998: 12). Species associated with the well watered stream valleys of the Cross

Timbers, include pecan, walnut, willow, elm, sycamore, and cottonwood (Story, 1985).

Along the Elm Fork of the Trinity in the Eastern Cross Timbers a mixture of ash,

blackjack oak, black willow, burr oak, elm varieties, hackberry, honey locust, mulberry,

pecan, post oak and wild plum are available (Story, 1985; Prikryl, 1990). Trees from a

variety of settings, including floodplains, uplands and stream valleys, within the Cross

Timbers provided a valuable food source, as well as, fuel for fire.

57

The Western Cross Timbers share many of the same characteristics as the Eastern

Cross Timbers separated only by the vegetation of the Grand Prairie. Stream

floodplains are dominated by various hardwood species and Juniper clings to the steep

slopes along Rivers. The natural vegetation is a savannah of oaks that follow the

stream channel or cluster across the landscape accompanied by an under-story of mid

and tall grasses. A variety of plant species create diversity with a patchy, mosaic

distribution across the landscape. Even with the wide variation in soils, the climax

understory vegetation is fairly uniform with the predominant grasses being little

bluestem, big bluestem, Indian grass, switchgrass, Canada wild-rye, side-oats grama,

hairy grama, tall dropseed and Texas wintergrass (Ressel, 1989; Prikryl, 1990).

In addition to the two main vegetation zones, Prairies and Cross Timbers, two other

additional vegetation habitats are significant in the region and are important to

understanding prehistoric settlement patterns in relation to plant resources. First, the

transitional zones between physiographic settings referred to as the ecotones. These

ecotonal settings contain plant species that are common to both physiographic settings

and there are some species that are exclusively found in these zonal interfaces. This

transitional zone effectively serves as an area where a number of species common to

woodlands and prairies would have greater accessibility. Another small, but important

habitat would be the riparian and floodplain zones. A large number of trees, shrubs,

berries and other vegetation types are located in this zone and are not found

elsewhere. This is due to the more mesic microenvironment of the riverbank and

58

floodplain in contrast to the more xeric terrace habitats. Here a number of species

uncommon or absent elsewhere would be available in a spatially limited area.

Investigations of McCormick et al. (1975) suggest that floral resources are equally

distributed over the uplands, slopes, and bottomlands. Many individual species occur in

more than one environmental setting (Prikryl, 1990: 12). For example, roots, such as

prairie turnips, were probably an important food source available in both the post oak

savannas and prairies (Martin and Bruseth, 1987: 128). Bottomlands are considered to

be the most significant part of the entire region in terms of availability of native wild

foods in these areas (Lynott, 1977; Ferring, 1990:17). Mast crops and many plant

resources available in the bottomlands of the Eastern Cross Timbers are also present in

the bottomlands of the Blackland Prairie. A potential for a large mast crop also exists in

the upland oak woodlands of the Eastern Cross Timbers (Prikryl, 1990: 12). In contrast

to the uplands of the Eastern Cross Timbers, the Blackland Prairie uplands are

grasslands. Upland areas typically provide grains yet “seeds of the dominant grasses of

the prairie uplands are small and were probably not valuable food sources for aboriginal

people” (Prikryl, 1990: 12). However, “it is suggested that some plants in the prairie

uplands, such as forbs, are usable foods” (Prikryl, 1990: 13). Some investigators

suggest, “while human consumption of upland acorns in the Eastern Cross Timbers was

important, these mast resources were also clearly important in sustaining fauna hunted

by prehistoric people” (Prikryl, 1990: 12).

Many of these resources are available in more than one season with summer and

fall having the highest number of species for potential use. The “floral resources that

59

are most abundant between July and September are mustang grapes, wild plums,

acorns and pecans” (Prikryl, 1990: 12). Pecans are available in the fall and throughout

winter. Acorns become available in September and are plentiful through February

(Prikryl, 1990). The most abundant plant resource are these mast (nut) crops, that

would have made this region very attractive to prehistoric people, deer, squirrels and

other animals. As such, it is expected that north-central Texas offered an abundance of

resources in the fall and winter. This suggests that the region was seasonally occupied

and probably accommodated tribal aggregation.

In prehistoric times five main groups of plants were collected and eaten by various

Native American tribes: fruits, roots, seeds, greens and nuts (Steward, 1938; Ferring et

al., 1978; McCormick et al., 1975). Examples of edible fruits include plums,

blackberries, yucca, prickly pear, mesquite, clove current, mulberries, persimmons,

hackberries and grapes. Roots utilized include sunflower, milkweed, cattail, bulrush,

thistle, wild onion, potato varieties and wild buckwheat. Pecan, black walnut with burr,

blackjack, post, and black oak nuts were also eaten. Seeds are often found in

association with the grasslands, Cultures in the region were aware of the edibility and

nutritional value of these species and probably enjoyed access to these resources.

Vegetation has a timeless relationship with geology, soils, and climate that helps

form a picture of the past natural landscape. Available vegetation resources at certain

intervals effected humans, animals, and the interaction therein. In order to learn more

about human behavior in relation to the landscape inferences are made based on

geology, soils, vegetation, faunal remains and artifacts found within site-catchment

60

areas. Vegetation is nourished by the geology, soil and climatic conditions in each

physiographic region. This same vegetation transforms this energy that it has derived

from various combinations of the ecosystem into the characteristic biomass that

sustained human and faunal prehistoric populations for thousands of years.

Faunal Resources in North-Central Texas Faunal presence is directly associated with the botanical resources described in the

last section. Fauna is culturally significant in that it served as a prehistoric food source

and is an indicator of climate, seasonality and landscape. Fauna, adapted to the

Eastern and Western Cross Timbers, Grand Prairie and Blackland Prairie, provided a

variety of food resources for prehistoric populations. Previous archaeological

investigations are the main source of information for gathering data about animal

resources that would have been available during these various time periods. Faunal

data is limited to certain locations where excavations were conducted. As such

inferences are also extended to the fauna that would be drawn to a given habitat.

Verrett and Bousman (1973) view the Eastern Cross Timbers as an economically

valuable location for base camps. A variety of hunting and gathering opportunities are

provided in this area,

Small groups of deer, buffalo, and other game animals could be hunted in the small grassland openings within the Cross Timbers. This forested section would have been ideal for trapping fur bearing animals such as fox, beaver and for hunting deer. Fish, mussels and riverine fauna could be exploited along the small Rivers and creeks typical of the area (McCormick, 1975: 16).

61

Major potential faunal resources include bison, white-tailed deer, antelope, turkey,

raccoon, squirrel and rabbit. Other mammals associated with the Eastern Cross

Timbers are cottontail, jackrabbit, armadillo, opossum, beaver, skunk and black bear.

Seasonally available resources include migratory birds, especially waterfowl which

feed along the aquatic margins, spawning fishes, amphibians and reptiles, which are

more active in warmer seasons (Ferring and Yates, 1998). According to sources cited

by Prikryl (1990:11) “320 species of birds and 43 species of migratory waterfowl have

been identified in the Cross Timbers and Prairies.” Most large game species tend to

aggregate in greater numbers within the bottomland forest zone. This would have

been especially true “during fall and winter when acorns ripen and fall and when some

late fruits like persimmons ripen” (Ferring and Yates, 1998: 12).

Lynott (1977:30) believes that it was not feasible for prehistoric people to exploit

the acorn crops of the Eastern Cross Timbers uplands because, combined with their

acidic content, the acorns are smaller in size than those of the bottomlands. However,

he notes that the acorn upland mast crops were important in sustaining many species

of animals that were hunted by prehistoric people (Prikryl, 1990: 12). Flora of the

uplands would have been important in sustaining the herds of herbivores, principally

bison and antelope, hunted by prehistoric people (Prikryl, 1990: 13). Information

provided by Goodrum et al. (1971: 527-529) indicates that the acorn yields are heavily

used as food by deer, raccoon, squirrel, wild turkey and quail. Squirrels and wild

turkeys feed on acorns, and these nuts may constitute about 75% of the diet of

squirrels (Prikryl, 1990: 12). Goodrum et al. (1971) propose that acorns make up as

62

much as 50% of the diet of white tailed deer, a species that was frequently hunted by

prehistoric people (Prikryl, 1990: 12). Evidence suggests that white-tailed deer

provided the largest percentage of meat consumed by prehistoric people in the region.

Deer would have been common in the bottomland forests of both the Blackland Prairie

and the Eastern Cross Timbers, and in the uplands of the Eastern Cross Timbers. Other

species that were drawn to the mast crop also would have occurred in these three

settings (Prikryl, 1990: 13). Although fewer useful species would have been available in

the prairie uplands, bison and antelope were present.

Presumably the most important large animals of the undisturbed prairies were

bison, antelope, bear, and perhaps mostly along the wooded fringes, white-tailed deer.

Bison are reported to have been present in north-central Texas during Late Prehistoric

times (Prikryl, 1990: 13). Historic accounts indicate that bison were also present in

great numbers on the Blackland Prairie in the Brazos River area during the eighteenth

century (Newcombe, 1986). On the basis of faunal data from archaeological sites,

Dillehay (1974: 180) suggested that there were “two periods in the past--6000/5000 BC

to 2500 BC and AD 500- AD 1200/1300--when bison were essentially absent from the

southern plains and the vicinity.” In relation to this he has proposed that bison were

present on the Southern Plains before 6000 BC and between 2500 BC and AD 500.

Lynott (1979), on the other hand, believes that there is a lack of evidence to indicate

bison presence in north-central Texas during these. His review of the Holocene faunal

evidence pertaining to the sites in north-central Texas and has concluded that bison

were never very common in the tall grass prairies, but after AD 1200 their numbers did

63

increase to a certain extent. The faunal record is mainly one of terrestrial animals.

While isolated finds are common, assemblages from well-defined depositional contexts

with established taphonomies are uncommon. Faunal remains that are most commonly

reported from aboriginal sites in north-central Texas consist of deer, cottontail, turtle,

and mussels. Yet it should be noted that many sites investigated in the region have

acidic sandy loams that do not preserve bone well. Furthermore, limited attention has

dedicated to collection and identification of animal remains until recently. As such,

“animals may have made up a more significant part of the aboriginal diet than evidence

indicates” (Prikryl, 1990: 13).

Excavations conducted along the Elm Fork have provided a wealth of faunal data

where change through time can be observed (Ferring and Yates, 1997). In order to

evaluate fauna that was typically utilized during prehistoric times excavated sites from

the northern and southern portion of the Elm Fork are evaluated (figure 6). Fauna

recovered includes a variety of aquatic species such as fish, frog, turtle and beaver.

Turtle is by far the most common from this group. Species classified in previous studies

as occupying grasslands/woodlands, grasses/wooded edges, woodlands/wooded edges

and wooded edges are grouped into ecotonal settings. These fauna include some

varieties of turtle, turkey, cottontail, birds, skunk, white-tailed deer and fox. White

tailed deer and cottontail are the most numerous represented in the archaeological

record. Grassland species include lizards, squirrel, rodent, gopher, rabbit, pronghorn

and bison. Pronghorn are the most numerous from this group. Bottomland and

woodland species such as opossum, gray squirrel, snake, swamp rabbit, spiny lizard,

64

armadillo, red fox, and raccoon are grouped into the same habitat category. Wolf,

coyote, dog, snake, and unidentifiable small, medium and large mammals are grouped

into the “various” category. Deer, turtle, mussels, rabbit and turkey are the most

common fauna reported. Mussels, often found at archaeological sites in the region,

drew populations to local stream beds. However, they have “low caloric yield and have

been viewed mainly as a dietary supplement” (Prikryl, 1990: 13). Aquatic and ecotonal

settings appear to be the most heavily utilized locales in relation to food procurement

situated primarily along the Elm Fork of the Trinity. However, one site in particular in

the northernmost portion of the study area indicates a notable utilization of bottomland

fauna and grasslands rather than ecotonal aquatic zones. Unfortunately, no equivalent

data sets are available for the East or the West Forks. Only one site contains definitive

bison remains in the form of “bison scapulas” (Lynott, 1975:126). According to Lynott,

data support that subsistence strategies along the East Fork of the Trinity were based

primarily on plants and animals from bottomland-riverine zones. The variety of

resources supported by the bottomland environment, primarily include “deer, raccoon,

opossum, rabbit, squirrel, and wild turkey” (Ferring and Yates, 1998: 12). Although

currently the data are lacking there was a wide variety of faunal resources available to

prehistoric populations.

Faunal evidence recovered from archaeological sites ranges from the Middle Archaic

to the second phase of the Late Prehistoric. A map depicts the distribution of these

sites where fauna was excavated (figure 6). A graph (figure 6a) provides a summary of

collective data from sample sites regardless of time period. The purpose of viewing the

65

data in this manner is to evaluate the range of potential wildlife in each area all other

things being equal. When change through time is considered the fluctuations can be

observed to determine what types of fauna were available during specific time periods.

A series of graphs (figure 6b-figure 6c) illustrate excavation sequences. This information

is evaluated to look for correlations with hypothesized climatic conditions and to explain

settlement patterns in relation to the habitats that provided resources through time.

Data available for the Middle Archaic is limited to 41DN102 situated along the northern

portion of the Elm Fork. General trends fluctuate through time and suggest the

utilization of resources from a variety of environments. In comparison with other time

periods bottomland resources are the most dominant during the Middle Archaic. In the

earlier and later levels grassland and aquatic resources are abundant. Ecotonal

resources also follow the same trend. The middle levels at this site contain bottomland

and grassland resources in order of abundance. This pattern supports the drier climate

that has been hypothesized during this time period. Comparatively Middle Archaic

populations utilized a wider variety of resources than any other time period.

Sites that contain evidence of Late Archaic occupation include 41DN26, 41DN346,

41DN381, 41CO144, 41CO141 and 41CO150. The geographical distribution of these

sites provides a fairly balanced coverage of the Elm Fork. In all but one of these

locations aquatic resources are fairly abundant yet ecotonal resources, although

fluctuating, are continuously dominant. Site 41DN381 is the exception where a wider

range of resources are being utilized as indicated by the abundance of the “various”

category. This evidence supports that the climate was relatively wetter than the Middle

66

Figure 6. Archaeological faunal distribution from excavated sites in study area.

0% 20% 40% 60% 80% 100%

41DN197

41WS38

41COL36

41DN346

41DN102

41CO144

41DN103

41DN99

41CO141

41CO150

41DN372

AquaticGrasslandsBottomlandsEcotoneVarious

Figure 6a. Fauna recovered from archaeological sites in study area (data summarized from Lynott, 1975; Ferring, 1997, 1998; UNT 1990-2001).

67

DN27DN26 DN99

0

50

100

150

200

1 3 5 7 9 110

50100150200250

1 3 5 7 9 11

DN372

0500

1000150020002500

1 2 3 4 5 6 7 8 9

DN102

05

10152025

1 3 5 7 9 11 13

DN346

0

10

20

30

40

1 3 5 7 9

0

50

100

150

1 3 5 7 9 11 13528

DN103

0

50

100

150

1 2 3 4 5 6

CO144A

0

50

100

150

1 3 5 7 9 11 13

CO144B

0

5

10

15

1 2 3 4 5 6 7

CO150A

0

20

40

60

80

1 3 5 7 9

CO150B

0

500

1000

1500

2000

1 2 3 4 5 6

CO150C

0100200300400500600

1 2 3 4 5

1268

1680 19852750 2910

2284

LATE ARCHAIC LATE ARCHAIC

LATE ARCHAIC

CO141

0

10

20

30

40

50

1 2 3 4 5 6 7

CO141B

0

100

200

300

400

1 2 3 4 5 61282759

LATE ARCHAICLATE PREHISTORIC II

LATE PRE ILATE PRE I I

DN381

0

50

100

150

1 4 7 10 13 16 19

LATE PRE I I LA LATE PREHISTORIC II

LATE PRE I / II LA

M IDDLE ARCHAIC

610

LATE ARCHAIC

LATE ARCHAIC

LATE PREHISTORIC II LATE PRE I I LA

LATE PREHISTORIC II

Figure 6b. Archaeological fauna and change through time (Aquatic=Blue; Grassland=Green; Woodland=Brown; Bottomland=Dashed Brown).

68

CO144B

0

10

20

30

40

50

1 2 3 4 5 6 7

DN27

050

100150200250

1 3 5 7 9 11

DN26

050

100150200250

1 3 5 7 9 11

DN372

0200400600800

1000

1 3 5 7 9

DN102

0

10

20

30

1 3 5 7 9 11 13 15

DN346

05

10152025

1 3 5 7 9 11

DN99

0

50

100

150

1 3 5 7 9 11 13528

DN103

0

20

40

60

1 2 3 4 5 6

CO144A

020406080

100

1 3 5 7 9 11 13

CO150A

010203040506070

1 3 5 7 9 11 13

CO150B

0

200

400

600

800

1 2 3 4 5 6

CO150C

0

100

200

300

400

500

1 2 3 4 5

1268

1680 19852750 2910

2284

LATE ARCHAIC LATE ARCHAIC

LATE ARCHAIC

CO141

0

20

40

60

80

100

1 2 3 4 5 6 7

CO141B

0100200300400500600

1 2 3 4 5 6 71282759

LATE ARCHAICLATE PREHISTORIC II

LATE PRE ILATE PRE I I

DN381

0

50

100

150

1 4 7 10 13 16 19

LATE PRE I I LA LATE PREHISTORIC II

LATE PRE I / II LA

M IDDLE ARCHAIC

610

LATE ARCHAIC

LATE ARCHAIC

LATE PREHISTORIC II LATE PRE I I LA

LATE PREHISTORIC II

Figure 6c. Archaeological fauna and change through time

(Ecotone=Red; Various=Green).

69

Archaic in the sense that more water resources would have been available and it would

have been possible for terrestrial wildlife that was associated with wetlands to be more

mobile. Overall, there appears to be a trend that reflects fluctuating climatic conditions

due to a dramatic increase of aquatic resources followed by a decrease that is equal in

intensity. The fact that ecotonal resources are the most abundant indicates a

propensity to utilize the resources from both the grassland and woodland environments.

Only two sites contain fauna that is associated with the first phase of the Late

Prehistoric (41DN99 and 41DN381). Due to the “fuzzy” boundaries this discussion will

combine the two phases of the Late Prehistoric with more emphasis on the second

phase. Late Prehistoric II sites include 41DN26, 41DN27, 41DN99, 41DN103, 41DN346,

41DN372, 41DN381 and 41CO141. During the second phase of the Late Prehistoric

there is an interesting trend that provides a contrast to the other time periods. There

are two situations taking place. One is an inversion from the previous patterns

indicated by a dramatic shift from resources located in the ecotone to resources from

various environments. The “various” category indicates a wider range of food resources

and higher degree of mobility. In the northernmost part of the area at 41CO141

ecotonal resources are dominant throughout followed closely by aquatic resources.

Both faunal groups follow the same bell curve trend and there is a slight presence of

grassland fauna as well. To the south 41DN99, 41DN103 and 41DN346 enjoy a healthy

presence of aquatic fauna that increases and decreases through time. Sites 41DN99

and 41DN103 both maintain a fairly substantial abundance of ecotonal fauna. Site

41DN346, a little further south, follows an opposite trend and shifts to utilization of

70

fauna from the “various” category. Approximately 20 km southwest, a cluster of sites

(41DN26, 41DN27, 41DN372 and 41DN381) contain evidence similar to the other sites

assigned to this time period but with a notable abundance in woodland fauna. A

decrease in ecotonal resources and increase of various habitat echoes the pattern at

site 41DN346. During the Late Prehistoric, population, territorialism, seasonal rhythms

and fairly stable climatic conditions factor into a wide range of subsistence strategies.

Climate of North-Central Texas

North-central Texas is located in a dramatic transition between regional climates.

The striking change in vegetation from “the East Texas deciduous forests on the

eastern margin of north-central Texas to the grasslands of the plains to the west is a

vivid reflection of this climatic transition” (Diggs et al., 1999: 28). The climate of this

region has a “seasonal, subhumid precipitation regime and a strongly seasonal, thermic

temperature regime” (Ferring and Yates, 2001: 16). For the most part, “summers are

hot and winters are mild except for brief periods of cold temperatures associated with

Arctic fronts” (Ferring and Yates, 2001: 16). These fronts frequently are “accompanied

by rain, and less frequently by snow and/or ice storms” (Ferring, 2001: 16). These

general trends are thought to be fairly consistent through time.

In terms of precipitation, there is a steep East-West gradient across north-central

Texas (figure 7). Mean annual precipitation ranges from about 46 inches in the

northeastern corner of the area to about 24 inches in the Western-most portion of the

Western Cross Timbers. In general mean annual precipitation decreases about 1 inch

for each 15 miles across Texas from East to West. This region is “prone to drought

71

Figure 7. Precipitation patterns in Texas and north-central Texas.

72

FicePaD

West Fork Climate Data (1951-1980)

010203040

J F M A M J J A S O N D

Month

Tem

pera

ture

(F

)

0

2

4

6

Prec

ipita

tion

(In)

Temp Precip

Elm Fork Climate Data (1939-1969)

010

2030

40

J F M A M J J A S O N D

Month

Tem

pera

ture

(F

)

0

2

4

6

Prec

ipita

tion

(In)

Temp Precip

East Fork Climate Data (1934-1963)

0

10

20

30

40

J F M A M J J A S O N D

Month

Tem

pera

ture

(F

)

0

2

4

6Pr

ecip

itatio

n (In

)

Temp Precip

gure 8. Average monthly temperature and precipitation trends in north-ntral Texas (compiled for West Fork (Ressell, 1983); Elm Fork (Ford and uls, 1978); and East Fork (Hanson and Wheeler, 1969) from Wise,

enton, and Collin County Soil Surveys respectively).

73

while at other times it receives too much rain in a short time” (Diggs et al., 1999: 29).

When the climate patterns occurring in the area around the West Fork, Elm Fork and

East Fork are compared (figure 8) the differences are minimal. The most notable

difference is the increased temperature and decreased precipitation, in the East Fork in

December. The opposite situation is true for the Elm Fork whereas both temperature

and precipitation are relatively low in the West Fork portion of the study area.

Combined Climate Data

0

5

10

15

20

25

30

35

J F M A M J J A S O N D

Month

Tem

pera

ture

(F)

1

2

3

4

5

6

7

8

9

10

Pre

cipi

tatio

n (In

)

Wise Temp Denton Temp Collin TempWise Precip Denton Precip Collin Precip

Figure 9. Comparisons of Average Monthly Temperature and Precipitation Trends in north-central Texas (compiled for West Fork (Ressell, 1983); Elm Fork (Ford and Pauls, 1978); and East Fork (Hanson and Wheeler, 1969) from Wise, Denton, and Collin County Soil Surveys respectively).

74

In general, precipitation and temperature data (1931-1989) indicate that late spring

and early fall are the wettest months while summer temperatures are high

accompanied by a decrease in rainfall (figure 9). These historical records indicate the

average rainfall as 32 inches per year and the average temperature is 65.2 F. The

similarities in climatic data for the three counties in the study area indicate that the

climate in general was fairly uniform with minor variations on the theme.

Past Environments

The entire eco-system is interdependent and certainly the climate was a major

factor for decision making in the past. Past environments have been extrapolated from

pollen and other organic data preserved by stratified profiles. The Ferndale Bog pollen

record is the sole source of early Holocene vegetational history for the region. Given

the proximity to the study area combined with the presence of these vegetation families

it is inferred that similar patterns likely occurred in north-central Texas during these

time periods. For the current research, this has been taken into consideration and

these data are mapped to provide a view of change through time in relation to the

relative abundance of these taxonomic families.

From a long-term perspective, pollen, plant macro fossils, and other types of

evidence demonstrate that the climate of Texas has changed substantially over the past

15,000 years (Diggs et al., 1999: 30). Paleoenvironmental change is not well

documented, but it is summarized by Prikryl (1993: 192-193) and updated by Ferring

and Yates (1997: 39-45). Prior to 14,000 BP the climate of north-central Texas was

cooler and moister than at present. Between 14,000 and 11,000 BP, the climate

75

became drier. It is suggested that the presence of high grass pollen and low arboreal

pollen between 7550-3050 BP shows a drying return of arboreal pollen after 3050 BP.

The latter change is similar to today’s environment. High grass pollen also occurs at

approximately 1550 BP and from 3550 to 3650 BP, and this also indicates drier periods.

The presence of paleosols between 2000 BP and 1000 BP suggests a decrease in

moisture with a return to wetter conditions after 1000 BP (Skinner and Ferring, 1999).

Prehistoric climatic data is in great need of improvement. Palynological data

supplemented by data from insects has the potential to add a tremendous amount of

breadth to the current research methodology and resulting settlement model.

This chapter has reviewed the geology, topography, hydrologic networks, soils,

physiography, vegetation, faunal communities, and general climatic trends that have

given the environment in north-central Texas its character over the past 10,000 years.

The environmental foundation is half of the entire ecological system. The other half of

this story is completed by the human populations that inhabited the landscape. A

chronological overview of these cultures provides a means to observe change through

time and establish a set of expectations regarding the range of behavior that has been

supported by previous archaeological evidence.

76

CHAPTER 4

CULTURAL CHRONOLOGY OF NORTH CENTRAL TEXAS

This chapter defines the characteristics of prehistoric populations, in north-central

Texas, as they change through time. Previous research has established specific

characteristics that define cultural subdivisions. Artifacts are a primary source of

differentiation between cultural groups and are utilized as an indication of time period

for surface sites that lack radiocarbon dates. Therefore, determining the range of

“diagnostic” projectile point types and character of the tool kit, for each time period, is

pertinent to the current research. Artifacts are a significant source of information

related to prehistoric activities. Subsistence evidence facilitates comparisons between

time periods and serves as an explanatory mechanism for site location or variable

density of occupation. Settlement pattern behavior, as discussed by previous

investigators, provides a means for comparison. These data collectively establish a

cultural framework to relate to the results of the SCA conducted on archaeological sites

dating from the Early Archaic to the Late Prehistoric.

North Texas prehistory spans across approximately 12,000 years. Geographic

distributions of prehistoric cultures reveal that ancient populations inhabited virtually all

of Texas with the exception of a few counties where evidence has yet to be reported.

Based on data provided by a variety of sources (THC, 1985; Skinner, 1985; Ferring,

1997, 1998; TARL 2002), a series of maps have been built to illustrate the change in

prehistoric Texas population distribution through time, with respect to physiographic

77

boundaries, the north-central Texas region and the research study area. The first of

these maps illustrates all prehistoric populations of record (figure 10).

Figure 10. Distribution of prehistoric sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

78

Chronologies that have been developed for north-central Texas vary according to the

investigator and reflect the lack of clarity or consensus in relation to cultural

subdivisions. These chronologies (table 1) have been developed and correlated with

projectile point typologies (figure 11). There is disagreement as to the correlation

between the precise time period, point typology and cultural subdivisions. However, in

general, there seems to be a certain level of agreement as to the general parameters of

the various time periods. This is a geographically related issue that depends on the

amount of work done in a certain region and is relative to few sites that have been

radiocarbon dated. The overlap between time period designations and the various

labels given to each time period by different investigators are important to note when

comparing archaeological reports and corresponding regional models.

Based on stratified excavations “point types” have been established as “index

fossils” to define each time period. Notable chronological sequences of point types that

have been developed in Texas (Prewitt, 1981; Prikryl, 1990; Story, 1990) have served

as the main reference for archaeological investigations. An overview of the currently

accepted point type chronology for this research is illustrated pictorially (figure 11).

These point types were recovered from archaeological sites, included in the current

research, and are associated with time periods based on radiocarbon-dated

investigations (figure 12). Archaeological data available for sites in north-central Texas

is contained in a variety of contract reports, academic publications, journal articles and

site reports filed with Texas Archaeological Research Laboratory (TARL).

79

Chronology

Archaic Neo-American McCormick et al. 1975

9000 BP-1500 BP 1500 BP- 450BP

Chronology Archaic

8000 BP-1200 BP Neo-American 1200 BP- 350BP

Skinner et al. 1985

Carrollton Focus

8000 BP- 4500 BP

Elam Focus 4500 BP-1200 BP

Wylie Focus 1100 BP-400 BP

Henrietta Focus 700 BP-250 BP

Chronology Early

Archaic Middle Archaic

Late Archaic

Johnson and Holliday, 1986:7

8000 BP- 6000 BP

600BP- 4000 BP

4000BP-2000 BP

Chronology Early

Archaic Middle Archaic

Late Archaic

Late Prehistoric I

Late Prehistoric II

Ferring et al. 1990

8500 BP- 6000 BP

6000 BP- 3500 BP

3500 BP- 1300BP

1300 BP- 800 BP

800 BP- 350 BP

Chronology Early

Archaic Middle Archaic

Late Archaic

Late Prehistoric I

Late Prehistoric II

Lebo and Brown 1990; Prikryl, 1990

8500 BP- 6000 BP

6000 BP- 3500 BP

3500 BP- 1250BP

1250 BP- 750 BP 750 BP- 250 BP

Chronology Early

Archaic Middle Archaic Late

Archaic

Late Prehistoric

Ferring et al. 1997

8500 BP-6000 BP

6000 BP- 3500 BP

3500 BP- 1250BP

1250 BP-350BP

Table 1. Various chronologies developed for cultural subdivisions.

80

Fpfr

igure 11. Overview of Early Archaic through Late Prehistoric I generalized projectile oint chronology in north-central Texas that is utilized for current research (compiled om Turner and Hester, 1993; illustrations used with permission).

81

Time Period

Early Archaic

Middle Archaic

Late Archaic

Late Prehistoric I

Late Prehistoric II

Morrill

Wells

Marshall

Tortugas

Palmillas

Carrollton

Castroville

Frio

Dallas

Ensor

Elam

Ellis

Trinity

Gary

Edgewood

Godley

Yarborough

Alba

Kent

Scallorn

Fresno

Bonham

Perdiz

Steiner

Washita

Catahoula

Maud

Toyah

Figure 12 Chronological overview of projectile point types from north-central Texas (data compiled from Suhm and Jelks, 1962; Turner and Hester, 1993).

82

Paleoindian (11,500-8,500 BP)

Archaeologists generally agree that the first people that migrated into Texas arrived

around 12,000 years ago around the end of the Pleistocene. These early inhabitants

left behind distinctive artifacts that reveal sophisticated tool manufacturing techniques.

Although Paleoindian artifacts have a widespread geographical distribution in Texas

most evidence is found on the surface (figure 13). In north-central Texas the only well

documented “in situ” excavation and subsequent publication, regarding a Paleoindian

culture in north-central Texas, is the Aubrey Clovis site (Ferring, 2001). Otherwise,

Paleoindian sites in the north-central Texas region are predominantly surface exposures

in mixed association with later occupations. The current research does not include

analysis of Paleoindians but it is necessary to review the characteristic artifacts, lifeways

and settlement patterns associated with these populations because they essentially set

the stage for the Archaic cultures.

Paleoindian cultures are defined by specific distinguishable characteristics that

differentiate them from all of the other time periods. The single most distinctive trait of

the Paleoindian cultures is the presence of unique and finely crafted projectile points.

Clovis, Folsom, Midland, Plainview, Firstview, Eden, Angostura, Scottsbluff, Dalton, San

Patrice and Golondria are projectile points (figure 14) associated with the Paleoindian

period. Of the forty-two projectile points, documented by Prikryl (1990): Clovis, Dalton,

Golondria, Midland, Plainview and Scottsbluff provide evidence of a Paleoindian

occupation in north-central Texas. The Paleoindian period has been defined by three

horizons based on projectile point typology: Clovis, Folsom, and Plano respectively.

83

Figure 13. Distribution of Paleoindian sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

84

The Clovis horizon is primarily represented by finely crafted points with basal flutes and

laurel-leaf shaped bifaces. Tool kits include a limited variety of tools consisting of

blades, scrapers, and bone or ivory cylindrical shafts (Lynott, 1981). It is common for

this time period that imported raw materials, rather than locally available quartzites and

cherts, were utilized in the manufacture of projectiles and tools (Ferring, 2001).

Utilization of non-local raw materials indicates increased mobility for long-distance

procurement by direct acquisition or evidence of cultural interaction.

Tool assemblages associated with the Folsom horizon include fluted and non-fluted

projectile point forms and various stages of lithic reduction that reveal a predominant

bifacial technology. Scrapers, choppers, perforators, hammerstones, abraders and bone

needles are also characteristic tools (Story, 1990: 189). Shifts in technology are

indicated by an increased frequency of unfluted points, such as, the Plainview, Midland,

and Eden. As with Clovis horizon artifacts, high-quality lithic raw materials were used

for Folsom projectile points, implying long-distance travel or trade.

Plainview, Eden, Meserve, Angostura, Scottsbluff, Dalton and San Patrice point types

are commonly associated with the Plano horizon. In correlation with other regions these

points indicate a date around 10,000 BP. Dalton and San Patrice points are associated

with the Paleoindian period. However, lifeways suggest affinities with the Early Archaic

period. Therefore these particular projectile points are considered to be part of a

transition between Paleoindian and Early Archaic technologies (Lynott, 1981:104;

Prikryl,1987:157). Given that many of these point types were discovered in the study

area, the Plano horizon is particularly important to the current research.

85

Figure 14. Paleoindian Projectile Point Chronology for north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).

86

Surveys conducted along the Elm Fork of the Trinity by Prikryl (1990) mainly recovered

Plainview and Dalton varieties. Dalton tool kits contain a variety of scrapers, retouched

pieces, unifacial tools, blades, blade-like flakes, bipolar cores, wedges and specialized

tools such as the uniquely fashioned bifacially chipped stone adze that was probably

employed for woodworking (Story, 1990:190-202). Unlike the Clovis and Folsom the

Plano horizon encompasses a more diverse set of cultural components as characterized

by a proliferation of projectile point types and tools.

Paleoindian cultures hunted extensively in a fairly generalized manner from very

small species to megafauna. Many Paleoindian sites contain remains of large mammals

such as: mammoth, mastodon, bison, caribou, deer, camel, and horse. Faunal analyses

at the Aubrey Clovis site indicate broad exploitation of large, medium, and small game,

including bison, deer, rabbits, squirrels, fish, and turtle (Ferring and Yates, 1997). The

range of variation presented by the faunal data indicates very flexible adaptive

strategies (Ferring and Yates, 1997). Many of the large mammals found in association

with Clovis sites became extinct requiring new adaptations. Folsom populations

adapted adjusting their lifeways to respond to the challenges presented by these

changes that are often attributed to the natural environment. This adjustment included

a shift to hunting herds of Bison, refined tools and new techniques. The Plano horizon

is believed to be the "final big game hunting occupation of the southern plains" (Lynott,

1981: 101). In north-central Texas, evidence is lacking for a subsistence strategy

based heavily on big game hunting. A generalized hunting and gathering lifeway is

considered more likely for Paleoindian populations that inhabited this region.

87

Vegetation during the Clovis horizon in this region was transforming due to

decreased rainfall and warmer temperatures toward the end of the Pleistocene. Clovis

point distribution suggests that all major environmental zones were exploited at one

time or another (Story, 1990: 178). Around 9,200 BP the Folsom horizon developed in

association to climatic changes and a complete shift to the hunting of bison. Shortgrass

prairie ecosystems developed during this time period and were preferred by bison

(Lynott, 1981: 101). During the Plano horizon evidence suggests that populations

associated with Dalton points utilized woodland edge and prairie environments whereas

San Patrice populations are associated with a more restricted range of environments.

In general, the Paleoindian period is characterized by a series of flexible adaptations

that correspond to the changing environmental conditions of the Late Pleistocene.

Paleoindian Settlement Patterns (11,500-8,500BP)

The settlement pattern for Clovis horizon populations is focused on upland settings

and along small tributary streams (Story, 1990:182). In general, Clovis populations in

north-central Texas were highly mobile and traveled considerable distances in small

bands (Lynott, 1981: 101; Johnson 1989), indicating an overall low population density.

Folsom presence in the region appears to have been very limited (Story, 1990: 189).

Towards the end of the Paleoindian era, it has been speculated that a slight reduction

of group mobility occurred in north-central Texas, possibly due to an increase in the

tallgrass prairies coupled with an increase in population density (Lynott, 1981:101-103).

However, repeated occupation of late Paleoindian sites by Early Archaic cultures

obscures whether or not the increased population densities can be attributed to Plano

88

horizon cultures. Nevertheless, the Plano horizon is mainly associated with Dalton points

which are fairly numerous in north-central Texas suggesting more intensive use of the

area in the Plano horizon than during the Clovis, Folsom or Early Archaic. Lithic raw

materials indicate that the settlement patterns of these groups focused on primary and

sometimes far-removed sources such as Alibates, Edwards Plateau, and the Ouachita

Mountains supporting a significant level of mobility (Prikryl, 1990; Banks, 1990). Data

from these sites combined with the lack of permanent structures indicate the existence

of small, highly mobile groups utilizing scattered campsites. Overall, higher population

densities, a reduction of mobility, and an emergence of territories are indicated towards

the close of the Paleoindian era (Story, 1990: 196). Although notably different in

character, hunting and gathering lifestyles endured throughout the Archaic.

Archaic (8,500-1,250 BP)

The Archaic time period represents a series of adaptations that were very similar to

the Paleoindians (McGregor, 1987). The geographical distribution of Archaic cultures

provides insight into the overall population density in Texas as it relates to the north-

central Texas region (figure 15). Patterns indicate a fairly widespread distribution and

increased population as suggested by the increase of sites reported across the state.

The Archaic period has been defined as a time period designation for hunter and

gatherer populations who practiced significantly different life-ways than their

predecessors in relation to mobility and subsistence (Story, 1990:211). As previously

mentioned, the varying chronologies produced by investigators for North Texas reflect

the lack of clarity as to cultural subdivisions of the Archaic. However, the number of

89

Archaic period subdivisions primarily reflects differences in the intensity and

geographical concentration of research conducted in each region (McGregor, 1987).

Figure 15. Distribution of Archaic sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

90

The more generalized chronologies reflect a lack of well-stratified sites and dated

components. One of the main difficulties with Archaic age sites is the tendency for the

mixing of cultural layers and frequency of surface finds that may be related to site

formation processes or re-use of lithic materials in areas of low sedimentation.

However, a handful of stratified sites have contributed to the development of

chronologies based on projectile point types found in radiocarbon-dated strata. The

current research depends on a chronology that has been most commonly accepted and,

in part, based on radiocarbon dating in north-central Texas. This provides a date for

the Archaic between 8,500 BP and 1,250 BP (Ferring and Yates, 1998).

Varieties of point styles (figure 16) that emerge during the Archaic provide a view of

the tremendous diversity that bloomed and brought the Paleoindian period to an end.

Rather than a small assortment of finely crafted artifacts, Archaic cultures adapted by

developing an extensive assortment of stone tools. In addition to a variety of

projectiles, scrapers, knives, gravers, burins, drills, gouges and axes were added to the

toolkit (Lynott, 1986; Story, 1990; Ferring and Yates, 1998). During the Archaic a

variety of small, stemmed dart points are first evident and polished stone tools appear

for the first time along with grinding implements, mortars, manos, pestles, and a variety

of bone tools (Ferring and Yates, 1998; Story, 1990). Hearths, middens and wells are

significant features that are heavily utilized during the Archaic (Johnson and Holliday,

1986; Story, 1990). Archaeological remains from this time period reflect clear changes

in cultural adaptations and increasing diversity in artifact assemblages suggest an

increased dependence on plants. The arrival of this diversity is probably related to

91

adaptations in response to population growth indicated by a substantial increase in

sites, increased knowledge regarding effective plant utilization and fluctuating climates.

Figure 16. Overview of Archaic projectile points in north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).

92

Early Archaic (8,500-6,000 BP)

The geographical distribution of Early Archaic cultures (figure 17) indicates a

population increase as compared to the Paleoindian. In north-central Texas, the Early

Archaic period is very difficult to define due to the lack of well-documented, temporally

unmixed cultural deposits and use of inconsistent artifact typologies (Story, 1990:217).

To date, few sites with discrete Early Archaic components have been investigated in this

region, with the exception of Prikryl's (1990) identification of 57 diagnostic Early Archaic

artifacts in assemblages from 22 surface sites in the Upper Trinity basin.

Compared to Paleoindian period artifact assemblages, Early Archaic (8,500 BP-6,000

BP) flaked stone tools tend to be less carefully fashioned and less often made of

extralocal raw materials (Story, 1990:213). Point types that define this time period vary

according to geographic location. Evidence of Early Archaic occupations in the Ray

Roberts Lake area consists of surface finds of the Angostura, Wells and early split

stemmed projectile point types (Prikryl, 1987: 158-161). The general evolutionary

trend in projectile point manufacture (figure 18) throughout the period is toward

shorter triangular points. Compared to the Paleoindian period, the tool kit of the Early

Archaic is more diverse. Projectile points are primarily produced from local cherts and

quartzites. Appearance of polished and ground stone items suggests changes in the

processing of food resources and most likely the intensification of plant food collection

(Story, 1981; Ferring and Yates, 1998). In general, the Early Archaic is characterized by

adaptations to the modern fluctuating climate and environment that emerged during

93

the Holocene. Cultural responses to these new ecological systems include a wider

variety of projectile points and more variable artifact assemblages.

Figure 17. Distribution of Early Archaic Sites in Texas and north-centralTexas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

94

Figure 18. Early Archaic projectile points in north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).

95

Middle Archaic (6,000-3,500 BP)

The geographical distribution of the Middle Archaic (figure 19) indicates a fairly wide

distribution with an overall decreased population density. The Middle Archaic is defined

essentially as a continuation of Early Archaic life-ways, with a few changes in adaptive

strategies. Middle Archaic sites are rare in this region (Ferring and Yates, 1997). In

the Elm Fork Trinity Valley, projectile points of this general age are found at fewer

localities than are points of any other age (Prikryl, 1990). The only “in situ” site of this

age, discovered in the study area, is 41DN102, the Calvert site (Ferring and Yates,

1997). Deep burial, dry climates and reduced occupation potentials are probably

important factors in the small number of sites dating to this period.

The growing diversity reflected in the material remains is the defining

characteristic that distinguishes the Middle from Early Archaic time periods. Generally,

projectile point size increases from the Early Archaic to Middle Archaic. In north-central

Texas, the Middle Archaic is characterized by an increased variety of projectile points

(figure 20). Dart points (e.g. Carrollton, Elam, Morrill, Wells and Gary); basal notched

projectile points (e.g. Andice and Calf Creek); and parallel, slightly contracting based

points (e.g. Pedernales, Bulverde, Travis and Nolan) are characteristic of this time

period (Prikryl, 1987: 162; Lebo and Brown, 1990). Regardless of the paucity of sites,

enough archaeological evidence has been discovered to formulate two cultural

subgroups referred to as the Trinity Aspect and Elam focus.

The Trinity aspect, which contains the early Carrollton focus, is one of the earliest

well-defined Middle Archaic period cultural entities centered along the Trinity. The

96

Carrollton focus is defined solely on the basis of artifacts and primarily projectile points.

The vast majority of the points are medium-sized to small in length. There is also

tremendous variation in the sizes of the individual point types such as the Gary point.

Figure 19. Distribution of Middle Archaic sites in Texas and north-centralTexas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

97

Figure 20. Middle Archaic projectile points in north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).

98

A few of the Carrollton focus projectile points are similar to Plainview and others

resemble the Gary types of the Late Archaic period. Other points associated with this

"focus" include the Carrollton and Trinity dart points (McGregor, 1988:31). The tool kit

includes spokeshaves, the "Waco sinker", the chipped stone Carrollton axe, scrapers,

gouges, a few grinding stones, ground stone, drilling tools, knives, hammerstones and

choppers (McGregor, 1988:31).

The most commonly recovered artifacts from Elam focus components are projectile

points. Possibly the most common projectile point of this period is the Ellis dart point,

followed by the Yarbrough and Elam types (McGregor, 1988:31). Similar to the

Carrollton focus, drills, scrapers, knives, hammerstones and choppers remain a

significant part of the tool kit. In comparison with the Carrollton focus, the Elam focus is

characterized by a higher percentage of lithic tools made from locally obtained quartzite

(McGregor, 1988:31). Overall, however, raw materials commonly used for Middle

Archaic tools derive from locally available lithic resources. Typology is complex and

often ambiguous requiring constant reevaluation. Nevertheless, the currently

established descriptions of Middle Archaic typological characteristics offer an adequate

overall cultural model for this time period.

Late Archaic (3,500-1,250 BP)

The geographical distribution (figure 21) of Late Archaic cultures indicates increased

population densities across the state. Significantly more Late Archaic sites have been

investigated in north-central Texas, than in previous time periods (McGregor and

Bruseth, 1987; Prikryl, 1990; Peter and McGregor, 1988; Ferring and Yates, 1997). The

99

abundance of sites is attributed to shallow burial below floodplains that has produced

in-situ sites with well-preserved living surfaces, features and biotic remains. These sites

have been found to be very common along the Trinity (Ferring and Yates, 1997:6).

Intriguing evidence for Late or Transitional Archaic use and occupation of the region is

related to large man-made depressions found at the Richland/Chambers Reservoir

(Story et al., 1990; Lynott, 1975). These depressions, referred to as "Wylie Focus pits,"

are features that may indicate group aggregation and ceremonialism (Lynott, 1975;

Bruseth et al., 1987; Story, 1990:235). Several sites excavated along the Elm Fork also

provide excellent stratigraphic records of Late Archaic cultures.

Most evidence for the presence of Late Archaic cultures (figure 22) includes

projectile points such as the Ellis, Edgewood, Elam, Gary, Dallas, and Godley varieties,

as well as, the slightly earlier Yarbrough and Trinity types recovered from

archaeological contexts (Prikryl, 1987; Brown and Lebo, 1991). Late Archaic

occupations in the Lewisville Lake area are primarily based on the surface recovery of

these point types with the exceptions of the excavations reported by Ferring and Yates

(1997). New developments in the material culture for this time period include the use of

pecked and polished stone in the production of ornaments and implements such as

spear-throwing weights and axeheads (Story, 1990:213). Prikryl’s (1990) analysis of

raw material types indicates that 39% of the Late Archaic tools were made from non-

local chert whereas Early-Middle Archaic and Late Prehistoric assemblages average

70%-80% chert raw materials. Intensive use of local materials is characteristic of the

Late Archaic period (Ferring and Yates, 1997:6).

100

Figure 21. Distribution of Late Archaic sites in Texas and north-centralTexas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

101

Figure 22. Late Archaic projectile point chronology for north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).

102

Archaic Climatic Change (8,500-1,250 BP)

The Archaic is a time period marked by climatic change from the end of pluvial

conditions through extreme xeric conditions (Johnson and Holliday, 1986: 46). These

conditions led to the modernization of faunal and vegetational communities. Beginning

in the Early Archaic period the gradual warming trend posited toward the end of the

Paleoindian period significantly changed during the early Holocene toward the

increasingly warmer, drier conditions of the Middle Holocene (Johnson and Holliday,

1986: 46). Although the trend towards a warmer, drier environment began in the Early

Holocene, wetter climates predominated during the Early Archaic (Ferring and Yates,

1997). Grasses were dominant between 9,000 and 5,000 BP (Ferring, 1993; Prikryl,

1987:156) effecting the options and related choices these early populations made.

Opening of new prairie uplands and retreat of hardwood forests to the floodplains is

related to these climatic conditions (Lynott, 1981:103). Drier climates and decreased

biotic resources are suggested during the Middle Archaic due to the presence of high

grass pollen and low arboreal pollen between 7500 and 3000 BP (Ferring and Yates,

1998:36; Skinner and Ferring, 1999:10). Beginning 5000 BP, the north-central Texas

environment underwent an increase in effective moisture and temperature that would

have affected the variety of available plants. A return to wet conditions toward the end

of this time period led to the beginning of the Late Archaic. Around 4,500 BP evidence

suggests that a migration of oak savannah may have replaced a large portion of the

grasslands (Prikryl, 1987:162). By 4000 BP the climate began to approach present

conditions. The environment, as indicated by paleoenvironmental evidence, obtained

103

from archaeological sites, experienced a slight shift to wetter conditions during the Late

Archaic period. Additional supporting evidence of wetter conditions includes the

development of the West Fork Paleosol (Ferring, 1986: 112). With wetter

environmental conditions, it is expected that the density of vegetal resources would

have increased, resulting in the expansion of the Cross Timbers woodlands. This would

have resulted in an increase of the plants utilized for human consumption and biomass

for game animals (Prikryl, 1987:170). The Late Archaic occurs during a return to

moister and somewhat cooler conditions, perhaps similar to those of today (Johnson

and Holliday, 1986: 47). However, climatically, the warm wet conditions continued for

another 1000 years. Cultural adaptations throughout the Archaic appear to correspond

with changes in climate. As such, it appears that the Archaic cultures were most

dramatically affected by the fluctuating climatic conditions.

Archaic Subsistence Patterns (8,500-1,250 BP)

Throughout the Archaic in north-central Texas economies were based on hunting

and gathering (McGregor, 1987). The subsistence pattern of the Early Archaic period is

very similar to that of the Paleoindian period. Early Archaic cultures were also nomadic

seasonal hunters except the annual range utilized by these cultures appears to be

considerably reduced. In addition, emphasis appears to have been placed more on

gathering a wider range of plant species within this annual range (Story 1981:144;

Brown et al. 1987:5-6). Extinct bison are present into at least the beginning of the

Early Archaic. During the Middle Archaic data indicates an increased modern bison

economy and a diffuse subsistence economy that included a wide variety of available

104

resources (Brown and Lebo, 1991:8). Faunal data from sites in north-central Texas

indicate that Late Archaic populations exploited a mix of prairie, forest and riparian

species (Ferring and Yates, 1997:6). According to data generated by excavations white-

tailed deer, rabbits, turtles and mussels are the most common (Ferring and Yates,

1997). In addition, mast crop resources would have been particularly desirable to these

early populations. Although the importance of hunting appears to be generally reduced

throughout the Late Archaic subsistence evidence indicates continued hunting of bison

and antelope in the upland prairie (Prikryl, 1990; Brown and Lebo, 1991). Procurement

of riverine faunal and floral resources in the prairie, Cross Timbers and ecotones

involved intensive exploitation of local bottomland resources and seasonal activities

(Lynott, 1981:104; McCormick, 1990). In some cases subsistence economy is focused

on the oak-hickory forests found along the floodplains of major drainages (Brown and

Lebo, 1991:7). As a result of a more diverse economy, site types within the Archaic

period are more varied than in the Paleoindian period. Site types include: open camps,

bison kills, flint quarries and flint caches (Story, 1990: 211). In general, Archaic

populations were hunter-gatherers that utilized a wide variety and abundance of wild

food resources.

Early Archaic Settlement Patterns (8500-6000BP)

It is probably due to wetter conditions during the Early Archaic that sites are

primarily found on stream terraces near primary water sources rather than bottomland

settings (Skinner and Ferring, 1999:9). According to Prikryl (1990) there are fewer

bottomland sites reported on the Elm Fork than during the previous period. However,

105

Lynott (1981) suggests that there was an increased emphasis on the use of bottomland

food resources (Skinner and Ferring, 1999:9). Subsistence patterns indicate a diffuse

hunting and gathering strategy on the floodplains and in the prairie uplands (Brown and

Lebo, 1991:7; Lynott, 1981:103). Given the various climatic interpretations the range

of settings occupied during this time period could be an indication of a fluctuating

climate or seasonal occupation in relation to drier or wetter conditions. The

aggregation of Early Archaic period populations appears to have been confined to small

groups of nomadic seasonal hunters who utilized briefly occupied hunting camps and

followed seasonal rounds that focused on a variety of resident and migratory animals

along the main stem of the Trinity and its larger tributaries. In general, previous

investigations indicate an Early Archaic concentration in the floodplains, bottomlands,

uplands and along major tributaries. Site locations are situated in both the prairies and

cross timbers. However, no trends or patterns have been identified, on a regional

scale, in relation to physiographic or topographic settings, to provide a clear explanation

for site location.

Middle Archaic Settlement Patterns (6000-3500BP)

The few Middle Archaic sites that have been located are on the first terraces above

the flood plains (Skinner and Ferring, 1999:9). These sites tend to be situated along

the Elm Fork rather than the smaller tributaries (Skinner and Ferring, 1999:9). Evidence

suggests that the subsistence economy was generally focused on the Oak-Hickory

forests found along the flood plains of major drainages while other sites along the Elm

Fork are located on terraces above stream floodplains (Brown and Lebo, 1991:7; Prikryl,

106

1990). Both cases indicate a direct correlation with major water sources that would

have been highly desirable during the drier climatic conditions. As scheduling of

hunting and gathering became more efficient, the size of territories became smaller

(Brown and Lebo, 1991: 8). During certain seasons, it was possible for small social

aggregates to form without exhausting local resources. Yet it would have been

advantageous for these larger groups to disperse during times of food stress (Lynott,

1981: 104). The bison hunt requires large areas of land, as well as, a strategic system

to find herds, corral and capture them. Seasonal scheduling of activities was necessary

to synchronize with the migratory patterns and cyclical abundance of herds. However,

this pattern has not been clearly illustrated in relation to site locations as they

correspond with the necessary settings that would fulfill scheduling requirements while

sustaining prehistoric populations.

Late Archaic Settlement Patterns (3500-1250BP)

During the Late Archaic there appears to be a notable shift of site location to

floodplains, first order tributaries and minor tributary streams (Prikryl, 1990: 74;

Skinner and Ferring, 1999:9). The development of the West Fork Paleosol toward the

end of the Late Archaic reflects a wetter environment that would have influenced the

range of settings appropriate for human occupation (Prikryl, 1990: 74; Brown and Lebo,

1991:8). Subsistence strategies indicate that a wide variety of settings (e.g. uplands,

riparian corridors, bottomland hardwood areas) were utilized. An expansion of the

Eastern Cross Timbers associated with the wetter environment would have provided a

larger mast crop for consumption by humans and animals (Brown and Lebo, 1991:8).

107

Evidence suggests that subsistence in the upland prairie continues to focus on bison

with some use of riverine faunal and floral resources (Brown and Lebo, 1991:8). Along

major drainages, the subsistence economies reflect increased efficiency in the use of

local floodplain resources (Brown and Lebo, 1991:8). This efficiency was probably

related to better scheduling and exploitation technologies (Brown and Lebo, 1991:8).

Although Late Archaic sites are distributed over a larger area and appear to have been

larger or more frequently occupied there is no evidence to supports that sedentary

villages existed during this period. Settlements appear to correspond with locations

that contain access to abundant resources, with favored sites being repeatedly used

over long periods of time (Brown and Lebo, 1991:8). A few sites contain clear evidence

of repeated occupations and sites with higher artifact densities are thought to indicate

repeated use as well (Ferring and Yates, 1997:6). Evidence suggests that camps and

habitation areas were not occupied year-round yet group movements appear to be

within particular territories (Story, 1981:146). In north-central Texas, the Late Archaic

settlement patterns indicate increasing evidence for the reuse of campsites and a trend

toward seasonal aggregation of populations when large quantities of food were locally

available (Lynott, 1981: 105). Higher site densities and increased cultural complexity

accompany the end of the Late Archaic.

Late Prehistoric (1250-250BP)

The classification of sites on a statewide scale combines the two phases of the Late

Prehistoric (as defined in north-central Texas) into one time period. Therefore the

generalized geographical distribution (figure 23) has been combined to form one overall

108

view. The appearance of the bow and arrow in the archeological record is the hallmark

of the beginning of the Late Prehistoric period (Story, 1990; Ferring and Yates, 1998).

Figure 23. Distribution of Late Prehistoric sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).

109

In addition, an increased dependence on plants led to the use of ceramics followed by

the development of agriculture and a relatively sedentary lifeway. Prehistoric

populations appear to have been receptive to these changes as suggested by the

technological changes indicated by the archaeological record.

Although, Prikryl (1990) and others divide the late Prehistoric into two phases, Late

Prehistoric I (1,250BP-750 BP) and Late Prehistoric II (750BP-350BP), based on

projectile point types, “the number of sites with discrete assemblages is very small, and

the great majority of the Late Prehistoric sites that he identified have points assigned to

both phases” (Ferring and Yates, 1998). In the later parts of this period, Plains Village

traits are found to be more common in assemblages from the Elm Fork Trinity (Ferring,

1986) while Caddoan traits are more common in sites from the East Fork Trinity

(Lynott, 1975; 1981). As a result, in Late Prehistoric times both temporal and cultural

factors contribute to archaeological complexity. Excavations of several Late Prehistoric

sites at Lake Lewisville and Ray Roberts Lake and other local reservoirs (Skinner and

Baird, 1985; Peter and McGregor, 1988; Lynott, 1975) provide examples of in-situ

archaeological sites (Ferring and Yates, 1997:7).

Scallorn, Rockwall, Catahoula, Bonham and Alba arrow point types (figure 24) are

diagnostic of Late Prehistoric I cultures (Prikryl, 1987: 133; Ferring and Yates, 1998).

Prikryl (1987) maintains that most projectiles are made of quartzite, while chert was

used more frequently to make arrows during the second half of the period. Signature

artifacts include smooth, flat-bottomed ceramic vessels; Gary dart points and the

appearance of small, expanding stem arrow points such as the Scallorn type and an

110

Figure 24. Late Prehistoric projectile point chronology for north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).

111

assortment of tools including drills, scrapers, and grinding implements (Story,

1990:217, 356). Ceramics associated with Late Prehistoric I cultures are often tempered

with grog and bone (Brown and Lebo, 1990: 8). The most common cultural features

encountered are hearths and probable roasting ovens along with less frequently

occurring storage pits, burials, and possible structural remains (Story, 1990:356).

Limited evidence suggests that this period in north-central and east Texas evolved

directly out of the earlier cultures associated with the Late Archaic period. However, it

is important to consider the possibility of secondary information diffusion and recycling

that may have occurred due to site reoccupation. Dating of these technological changes

is speculative at best since a considerably earlier dating for the introduction of ceramics

has been suggested for portions of the middle Trinity drainage and for east Texas.

Late Prehistoric II occupations, in north-central Texas, are signified by Washita,

Harrell, Toyah, Maud, Perdiz, Steiner and Fresno arrow points (figure 24). Material

culture exhibits a distinct change in the composition of lithic tool assemblages. Other

material evidence includes bone tools; scrapers; edge-beveled knives; drills and shell-

tempered pottery (Prikryl, 1987; Brown and Lebo, 1990). Ceramics, projectile points,

and other material culture remains from archaeological contexts in north-central Texas

appear to be related to sites in eastern Texas from this same period (Story, 1981).

Introduction of the bow and arrow probably brought about a tremendous change in

hunting technique and efficiency, probably leading to a higher abundance of food

resources and a resultant increase in population. The shift from dart points to arrows

also implies adoption of individual hunting strategies over group hunting strategies.

112

Late Prehistoric Climatic Conditions

Around 1060 BP it is posited that the end of the West Fork Paleosol indicates a

return to a drier climate (Brown and Lebo, 1991:9). The presence of Paleosols between

2000BP and 1000 BP suggests a decrease in moisture during this period with a return

to wetter conditions after 1000 BP indicated by an increase in arboreal pollen (Skinner

and Ferring, 1999:10). The fairly steady decrease in the ratio of nonarboreal pollen to

arboreal pollen inferred from the Ferndale Bog Sample reveals a general trend toward

less dry conditions (Story, 1981). High grass pollen also occurs around approximately

1500 BP and during an interval between 500 to 400 years ago indicating that drier

climatic intervals probably interspersed with the wetter conditions that are suggested by

carbon isotope evidence (Ferring and Yates, 1998; Skinner and Ferring, 1999).

Late Prehistoric Subsistence Patterns

While changes appear to have occurred in most aspects of cultural lifeways, in the

Late Prehistoric, subsistence patterns changed little from Late Archaic times. In

general, evidence suggests that exploitation of riverine resources continues and the

exploitation of bison increases after A.D. 1200. However, throughout the Late

Prehistoric deer are considered to be the primary source of meat protein (Lynott, 1981,

Skinner and Baird, 1985; Ferring and Yates, 1997). Smaller game that supplemented

the meat portion of the diet predominantly includes rabbits and turtles. Plant foods

played an increasingly larger role in subsistence as indicated by the increase of

groundstone artifacts. Evidence from many sites in north-central Texas suggests that

the most common plant food source is the hickory nut (Martin, 1987) that would have

113

required grinding techniques in order to facilitate preparation for use. No tropical

cultigens have been identified in floral assemblages, although squash and sunflower

seeds are represented (Story, 1981: 146). Use of domesticates is indicated at several

sites (Peter and McGregor, 1988), but is not presumed to have been a dominant

economic focus. Corn horticulture prior to A.D. 1200 has only been found at a few

sites. As Story (1981: 146) proposes, subsistence was probably based on an intensive

foraging strategy and possibly included cultivation of native food resources. Ceramic

vessels are also an important innovation that may reflect alternative food preparation

techniques unknown during Archaic times (Story, 1981:146). This also may indicate the

beginning of an intensified subsistence strategy that may help explain reduction of

group mobility and population growth (Bruseth and Martin, 1987:284). In general, it is

hypothesized that Late Prehistoric cultures participated in a nomadic broad-spectrum

subsistence economy that included a primary focus on deer and seasonal scheduling.

The primary difference between the first and second phases of the Late Prehistoric

is related to climatic conditions. A more xeric climate is believed to have been a

dominant influence during the Late Prehistoric I era (Lebo and Brown, 1990). As such,

there is no indication of bison at this time. During the Late Prehistoric II period the

presence of bison remains at archaeological sites is considered to be supporting

evidence of the return to moister climatic conditions (Dillehay, 1974; Lynott, 1979;

Brown and Lebo, 1990). There is strong evidence from at least two archaeological sites

identified in Dallas and Denton counties that suggests a shift back to bison hunting and

a bison-oriented economy during the second phase of the Late Prehistoric time period

114

(Story, 1990:359; Ferring and Yates, 1997:6). One of these sites yielded two bison

scapula hoes and a tibia digging stick, while the other produced of eight bison scapula

hoes (Story, 1990:359; Ferring and Yates, 1997:6). Bison scapula hoes suggest a

combined economy of bison and rudimentary horticulture (Brown and Lebo, 1990).

Late Prehistoric Settlement Patterns

During the Late Prehistoric period site locations are very similar to the Late Archaic

in that they are situated along major tributaries such as the Elm Fork (Skinner and

Ferring, 1999:10). Sites are often found in the floodplains of area streams, especially

those of tributary streams. Strategic site selection is indicated in that sites were once

again located on sandy terraces above the floodplain in optimum settings.

Settlement patterns during the Late Prehistoric I era were effected by climatic

changes and technological adaptations. Important changes include shifts in settlement

patterns to larger sites that contain features including possible house structures.

Evidence such as the introduction of the bow and arrow suggests that there was an

external influence (Brown and Lebo, 1991: 9). More permanently settled villages are

expected related to the presence of maize. The transition from hunting and gathering

to a fully sedentary lifestyle was gradual in north-central Texas, probably beginning

during the Late Archaic but not established until the Late Prehistoric (Story, 1981:146).

Late Prehistoric II evidence suggests a shift in subsistence and settlement patterns

of the previously nomadic bison hunting groups. During this time period subsistence

became focused on riverine habitats with more sedentary settlements situated along

the major drainages (Brown and Lebo, 1991:9). Increased sedentism resulted in more

115

focal subsistence strategies (Brown and Lebo, 1991:9). These adaptations are thought

to be the result of internal population growth and external population pressure (Brown

and Lebo, 1991:9). Use of horticulture economies added another dimension to

subsistence opportunities (Brown and Lebo, 1991:9). Bison appear to have been

hunted on a more opportunistic basis rather than by the previous continual strategic

pursuit (Brown and Lebo, 1991: 9). Increased sedentism indicated by site size and

artifact assemblages is related to more focal subsistence strategies. Accordingly, in

north-central Texas, most of the Late Ceramic period sites, affiliated with the Caddo,

are concentrated at the eastern edge of the Eastern Cross Timbers. Over 80% of all

sites dating to this period along the Elm Fork of the Trinity are found at the prairie-

forest boundary (Prikryl, 1990:82). This concentration of settlement is hypothesized to

indicate responses to drier climatic conditions. “Neo-American” groups, who are

contemporaneous with Late Prehistoric cultures, followed a similar seasonal exploitation

pattern until agriculture became an important part of their subsistence base. In

general, during this time period the advent of agriculture led to the establishment of

large base camps or permanent villages that were typically located in or adjacent to

large standing alluvial bottoms of the more permanent water sources.

Unanswered Questions

Previous investigations have documented archaeological sites and formulated

theories concerning prehistoric settlement patterns in north-central Texas. Based on

these previous investigations it is possible to evaluate material culture, subsistence

strategies and settlement patterns as they change through time. It is also possible to

116

identify the gaps and questions that remain unanswered. While archaeological remains

are a known there are many unknowns that remain regarding how the people

interacted with the landscape and each other. Clues from prior investigations need to

be assembled into a coherent whole, gaps identified and new hypotheses tested, in

order to view the related behavioral patterns. Some of the unanswered questions can

only be answered by future research. In the meantime, the current research is

designed to synthesize and evaluate the results of previous investigations and formulate

hypotheses to fill in these gaps, where possible, answer some of these questions and

facilitate the construction of a regional model of prehistoric settlement pattern behavior

in north-central Texas. In order to accomplish this task it is necessary to construct a

unique methodology that has multi-scalar flexibility and a strong theoretical foundation.

117

CHAPTER 5

RESEARCH METHODOLOGY

Site catchment analysis (SCA) is an effective means to address hypotheses

formulated to answer the variety of questions posed by previous investigations and to

develop regional settlement models of prehistoric socio-economic behavior. Given the

multi-facetted character of the landscape, it is necessary to look for ecological

relationships that provide a strong economic resource base rather than one generalized

characteristic in isolation. This extends beyond physiographic settings, to include soil

interpretations, hydrological, geological and topographical relationships.

Ultimately the characteristics related to site location are the primary variables

correlated with site function, site type, occupation frequency, site interaction and

chronological relationships. The series of inter-connected relationships contribute to the

overall ecological structure. These relationships are evaluated to determine the

combinations of environmental features that influenced the choices and preferences of

prehistoric populations which are reflected by the corresponding settlement patterns.

The dynamic character of ecosystems requires that the inter-relationships of the various

components are evaluated separately in relation to a variety of factors and then joined

back together. Therefore it is necessary to test multiple hypotheses from a variety of

angles in order to gain a deeper understanding of how these components work together

to form a regional picture of prehistoric populations developing through time.

Essentially where a site is situated on the landscape is the primary geographical

foundation for SCA. Based on the location of each archaeological site and the character

118

of the artifact assemblage, a variety of research hypotheses are designed to evaluate

previous settlement pattern theories to answer the questions that prior investigations

left unanswered. In north-central Texas previous research has primarily evaluated site

location with respect to resource concentration in relation to physiographic settings,

hydrological patterns and topographic positioning.

For the current research, SCA is conducted on a regional scale to define the

character of 1km site catchment areas around archaeological sites. Sites are

collectively evaluated to formulate a regional model of prehistoric settlement patterns in

north-central Texas. SCA is performed by depiction, extraction, measurement and

evaluation of environmental data in correlation with archeological remains.

Environmental variables are derived from modern habitat descriptions. Data are

assembled on landforms (topographical, geological, hydrological) and potential of soil

types to produce economically advantageous (edible, medicinal, supplemental) plant

communities tabulated. Cultural variables are based on ethnographic analogy supported

by the evidence of the artifact assemblages collected from Texas Archaeological

Research Laboratory (TARL) files and a variety of reports. Criteria considered as viable

explanatory mechanisms for site selection and related settlement patterns include:

1. soil settings (taxonomic, textural, topographic) 2. soil potential to provide suitable habitat for plants and animals 3. economic value (edibility, medicinal, supplemental) of soils 4. catchment diversity 5. geological resources (surface geology) 6. geomorphology (uplands, bottomlands, floodplains) 7. physiographic resources (prairies and Cross Timbers) 8. climatic variables 9. cultural factors (archaeological composition and character)

119

Most of these criteria are mapped by utilizing digital data layers derived from several

sources. Some data layers did not exist prior to this research and had to be generated.

Conceptual criterion such as cultural factors, climatic variables and catchment diversity

are incorporated into the discussion rather than depicted on a map.

Pre-existing environmental digital data layers consist of County boundaries

(NCTCOG), geology (USGS), soils (SSURGO), streams (NCTCOG), Rivers (TNRIS), lakes

(NCTCOG), elevations (USGS) and physiographic regions (TNRIS). While the pre-

existing digital data layers are highly accurate it is important to note that there may be

slight inconsistencies, distortions or misalignment of the data layers, depending on the

scale of the data layer combined with the source that created the data. For the most

part, the data layers are seamless, consistent and accurate, especially those

constructed as a regional dataset by the same group of investigators. However, in the

case of the soils data layers there are slight inconsistencies in soil series classifications

that create seams at the County boundaries where proximal soils were categorized into

different soil series classifications. This is due to the fact that different investigators

conducted the soil surveys at different times and any slight categorical variation,

including utilization of a different name for the same soil type, will be enhanced when

the data is in digital format. In some cases it has to do with the level of detail and in

others the discretion of the particular soil surveyor. These differences are not

significant enough to affect the validity of the SCA but are brought to the reader’s

attention so that they may be aware of possible inconsistencies in the digital data

layers. This type of phenomena is taken into consideration in the current research and

120

should be addressed in any research that involves digital data layers created by

different sources.

Generated data layers include archaeological sites, site catchment boundaries,

economic values (edibility, medicinal, supplemental), topographic settings, riparian

corridors and ecotones. Environmental modeling of these data results in the mapping

of the distribution of topographic settings, physiographic settings, geological structure

and hydrological resources, plant and wildlife potentials and soil economic values.

While over the past tens of thousands of years there have been shifts from forests

to prairie environments in north-central Texas palynological and climatic evidence

indicates that over the past 8,000 years these changes are not extreme enough to

abandon inferences about prehistoric environments based on modern environmental

attributes. As such, modern soil types combined with their associated plant and animal

communities are a viable source for reconstructing prehistoric adaptive strategies. The

geological structure that forms the foundation for these soil types is stable enough to

support the development of similar soil types through time. What differs is the degree

of soil development and the climate that would have either facilitated abundant or

sparse resources related to those particular soil types.

A conceptual diagram (figure 25) provides an overview of the links and the data that

are collectively analyzed, in relation to archaeological site distribution, to shed light on

prehistoric settlement pattern behavior. The procedures used to perform SCA include

data collection, construction of databases, classification, scoring, data layer generation

and a variety of GIS techniques. GIS software also provides an interface to organize,

121

Figure 25. Conceptual diagram of data included in SCA.

Upland

Terrace

Floodplain

Geomorphology

Ecotones

Riparian Corridors

Woodlands

Grasslands

Soils Geographic Setting

Subsidiary

Medicinal

Edibility

Economic

Interpretations

Great Group

Soil Setting

Relationships

Rangeland

Wetland

Openland

Herbaceous

Grasses

Grains

Vegetation

Potential

Soils/Vegetation

Geology

Water

Supplemental

Medicine

Subsistence

Wildlife

Vegetation

Soils Potential

Climate

Geomorphology

Interpretations

Archaeology

Artifacts

Features

Choices

Prehistoric Settlement Patterns

Vegetation

Geology

Hydrology

Topography

Bottomland

Ridges

Wildlife

Site Catchment

Resources

Fauna

PhysiographyPhysiographic

Ephemeral Streams

122

manage, analyze and extract data to perform a variety of spatial and statistical

analyses. Terminology related to these GIS techniques is defined in a glossary provided

at the end of the appendix. GIS procedures are designed to conduct SCA and prepare

the data to address a variety of research hypotheses that collectively serve to build a

model of prehistoric settlement patterns.

Currently there are no known SCA studies that have set a precedent for the

particular approach created for this research. The current methodology is innovated to

respond to a series of questions regarding cultural affinities, contemporaneous

environmental opportunities, activities and subsistence strategies. Theories based on

past and present research are also tested to determine whether or not they are

applicable on a regional scale.

Soils as a Diagnostic Tool

SCA involves reconstructing the natural environment in past tense. The most

complex part of environmental reconstruction is associated with vegetation. In this

particular study soil is selected as a primary diagnostic attribute for the vegetation. As

a result a number of different investigators have employed a variety of techniques to

accomplish this goal including use of pollen diagrams, field methods, General Land

Office Surveys (GLO) and other historical surveys (Tiffany and Abbott, 1982: 314).

However, in this research, vegetation reconstruction is primarily based on soil potential

as inferred by soil survey data. According to Tiffany and Abbott (1982) the “systematic

changes in soil properties across a toposequence serve as proxy data for detailed

vegetative reconstruction” and “descriptions of soil types state the past vegetation

123

responsible for the formulation of that particular soil type” (Tiffany and Abbott, 1982:

316). Various SCA studies have employed soils as an explanatory mechanism. Soils are

utilized in the current research because they provide the foundation for the potential of

vegetation. Geology and soils are the most stable components of the environment.

Geology is the most stable as it is not subject to run-off and it is the parent material for

the soils. As such if there is increased rainfall that contributes to erosion the soil

formation process replenishes the ingredients of the soils that are subject to run off.

The character of the geology ultimately determines the character of the soil that

provides the matrix of possibilities for specific types of vegetation to exist. For

example, sandstone tends to produce sandy soils that lead to forest vegetation and

limestone tends to produce soils that are high in clay content creating an appropriate

environment for prairie vegetation. Any variation within the geology will lead to

variation within the soils which in turn leads to variation within the vegetation and this

is what gives the environment its diverse character. This variation is described by soil

taxonomy.

Climatic patterns in combination with topographic settings explain the abundance

and spatial distribution of specific types of vegetation that exist in response to the soils

and geology. Based on a sample of plant types, identified in the soil survey, research is

performed to determine plant edibility, medicinal value and supplemental value.

Economic potentials of soils are determined by soil survey correlations of soils to

vegetation according to settings where particular soils are typically situated. This

provides an interpretable view of potential prehistoric habitats. While the current

124

research is highly dependent on soils to ascertain economic value other environmental

variables are included to provide a more comprehensive view.

Soils Variables

Soil surveys contain an abundance of environmental information. Although many

descriptive attributes are attached to soil types the particular attributes that are used

for this study are limited to soil series, great groups, plant type potentials, wildlife

potentials, soil settings and topographic settings. Soil series are consolidated or

recoded to represent each of these variables. Pre-existing soils data layers exist for

each of the counties included in the regional study area (the West Fork or Wise County,

the Elm Fork or Denton County and the East Fork or Collin County). The only two

comparable soil surveys from the study area that are amenable to detailed analyses are

from Wise (West Fork) and Denton (Elm Fork) counties. Soils data for the East Fork of

the Trinity are limited to great group. As a result there are no equivalent correlations

with vegetation. Any relationships would have to be inferred. Although “great group”

scale soils data provide a means of comparison between the environmental contexts of

the East, Elm, and West Forks, the character of the environmental setting around the

East Fork is not suited for comparative analyses. The East Fork is clearly a significantly

different environmental setting than the Elm and West Fork of the Trinity. As such it

does not make sense to statistically evaluate the three settings comparatively.

Therefore, the East Fork is evaluated in a similar manner where possible and included in

comparative discussion but not statistically compared. Due to the lack of data currently

available statistical analyses involving the relative potential for vegetation types, wildlife

125

habitat and economic potential of site catchment areas are limited to the Elm Fork and

the West Fork of the Trinity.

Site catchment areas are composed of various scales of proportional values. Soils

data can be consolidated into soil series, order, settings or great groups. During the

organizational process of soils data it is important to note that a great group can consist

of several soil series or settings and likewise soil settings can include several great

groups. Soil series, however, are associated with one particular great group.

Therefore, the manner in which the data is consolidated is dependent on the research

question. For the current research the most appropriate approach is to conduct the

initial analysis on a soil series level and then consolidate the data into larger categories

when necessary. In order to recreate the variable potentials while considering soil

variability a soil series level approach is appropriate and preferable. This facilitates the

detection of variation within the great group while calculating plant potentials, wildlife

potentials, soil setting, topographic setting and catchment diversity. For the economic

analysis it is also necessary to work primarily from a soil series level and then

consolidate to soil settings and great groups to summarize the results.

Soil Settings

Soil series data is connected to attributes that describe taxonomic, topographic and

textural settings typically associated with certain soil series. These settings are

depicted by adjusting the symbology of pre-existing soils data layers in ArcMap.

Topographic settings associated with soil types include alluvial terraces,

bottomlands, floodplains, low hills and ridges, old terraces and valley fills, plains,

126

Pleistocene terraces, ridges, stream divides, stream terraces, terraces, upland ridges

and uplands. These settings are often referred to in the literature. While this provides a

sketch of the topographic setting the interpretive value of the particular setting is

increased when related to climatic conditions, physiographic settings, economic values,

related soils and faunal assemblages. Research conducted by Ferring and others

suggests that prehistoric populations primarily selected sites located in terraces and

upland settings in ecotonal transitional zones between prairie and woodland settings.

Hayes classifies sites on the Elm Fork into floodplain, terrace and upland settings.

According to previous investigations, during the Early Archaic, sites are situated on

stream terraces and in bottomland settings. Given the wet climate during the Early

Archaic it is expected that populations were concentrated in upland settings near

ephemeral streams. Given the dry climate during the Middle Archaic it is expected that

populations were concentrated in the bottomland, floodplain and riverine settings

associated with major tributaries. During the Late Archaic moister conditions are

expected to have increased the availability of plants for human consumption and

biomass for game animals. Vegetation retreated into the floodplain while drier

conditions prompted a return to the stream banks. Evidence also suggests utilization of

upland prairie settings. However, in general, subsistence strategies appear to have

been focused in riverine and bottomland settings within the prairies and Cross Timbers.

SCA techniques are designed to evaluate this combination of factors in relation to

occupation frequency and change through time. Patterns are also considered

comparatively between the Elm, East and West Forks of the Trinity. In order to gain

127

insight on these patterns hypothesis testing includes topographic settings in discrete

statistical analysis to evaluate site location with respect to occupation frequency and

change through time.

Soil textural settings include blackland, clay loam, clayey bottomland, claypan

prairie, deep Redland, deep sand, eroded blackland, loamy bottomland, loamy prairie,

loamy sand, low stony hill, Redland, rocky hill, sandstone, sandy loam, sandy, shallow

clay, shallow, steep adobe and tight sandy loam. These groups are primarily utilized in

the economic study to provide a linking mechanism between the vegetation and the soil

series. Great groups that are included in the study area (table 2) are evaluated in both

multivariate and discrete statistical analyses.

WEST FORK AND ELM FORK EAST FORKGREAT GROUP GREAT GROUP ARGIUSTOLLS CHROMUDERTS CALCIUSTOLLS CHROMUSTERTS HAPLUDERTS OCHRAQUALFS HAPLUSTALFS HAPLAQUOLLS HAPLUSTEPTS HAPLUSTOLLS HAPLUSTERTS PALEUSTALFS HAPLUSTOLLS PELLUSTERTS PALEUSTALFS RENDOLLS PALEUSTOLLS USTOCHREPTS RHODUSTALFS USTIFLUVENTS

Table 2. Soil taxonomy classifications. Soil series are consolidated into these great groups based on links established by the

soil survey. In order to attach economic value to these settings a “dissolve” operation is

performed rather than adjusting symbology. This allows the economic values to be

summed in relation to the consolidation of soil series into soils settings or great groups.

The relationship between the economic value and the particular soil setting is important

128

to the current multivariate analyses. For other parts of the analysis the archaeological

data and soil settings are evaluated to facilitate discrete analysis. In order to create

contingency tables ArcMap tools are used to select the particular settings by “attributes”

and archaeological sites (grouped into occupation frequency and according to

chronology) by “location” in relation to the particular selected settings.

Soil Potential for Plants and Wildlife

Plant types and wildlife potentials are associated with soil series based on soil

survey estimates within the overall study area and within each site catchment area.

These attributes are recoded (table 3) from soil survey ratings that were “high”,

“medium”, “low” and “very low”. Based on the frequency of “very low” categories for

these types of wildlife habitat, “low” and “very low” were combined. Soils that were

considered to have a “high” suitability for each type of habitat were given a value of

0.64. The soil types with “medium” potential were given a value of 0.29 and the soils

with “low” potential were given a value of 0.07. These weighting techniques are

derived from “Ranking Procedures” in GIS and Multicriteria Decision Analysis

(Malczewski, 1999:180).

Rank Sum Rank Reciprocal Rank Reciprocal Soil Rank Weight Normalized Reciprocal Normalized Weight Normalized

Suitability r = rank n = # of categories Weight Weight Weight Weight

r (n-rj+1) Weight/ Total Weight (1/rj)

(n-rj+1) p, where p=2

HIGH 1 3 0.50 1.00 0.55 9 0.64MEDIUM 2 2 0.33 0.50 0.27 4 0.29LOW 3 1 0.17 0.33 0.18 1 0.07Total 6 1.00 1.83 1.00 14 1

Table 3. Weighting procedure for plant and wildlife soil survey estimates.

129

Essentially this procedure transforms discrete into continuous variables so that they

may be proportionately assigned a value and evaluated in relation to their relative

representation within site catchment areas. These weights are attached to soil series

digital vector data by utilizing ArcMap to perform an “attribute join” operation. Once

the attribute data is joined to soil series location, with GIS tools, the spatial distribution

of these factors can be included in a multi-scalar spatial and statistical analysis.

Resulting data layers provide an overview of the spatial distribution of wildlife and

vegetation potentials for the entire County surfaces. In relation to archaeological sites

these data layers can be “clipped” based on the extent of site catchment areas with

ArcMap geoprocessing tools. Associated values that remain within the site catchment

territory can then be measured, described and statistically evaluated.

Interpretations: Potential Economic Value of Soils

Locally available botanical resources that could have been utilized by prehistoric

populations are of primary importance to SCA and settlement pattern investigations.

Although, ethnographic botanical information is sparse, a variety of publications have

been written that provide extensive information on the edibility, medicinal, and toxic

properties of plants that are native to Texas. In 1999, a comprehensive study

documenting the flora of north-central Texas compiled by Diggs, Lipscombe and

O’Kennon, was published. Tull (1987) put together a guide that discusses edible and

useful plants in Texas and the Southwest. Duke (2000, 2002) has written a number of

books that provide information about medicinal plants and herbs in the United States.

McCormick, Filson and Darden (1975); Skinner and Baird (1985); Lynott (1977); Martin

130

and Bruseth (1987); and others also provide information regarding the potential of

botanical resources in north-central Texas. According to Prikryl (1990), one of the best

publications concerning floral resources of the lower Elm Fork area is that of

McCormick, Filson and Darden (1975), who studied the terraces of an area where

Timber Creek enters the Elm Fork floodplain. Situated on the boundary of the Eastern

Cross Timbers and Blackland Prairie, this locale provides a healthy representative

sample of the vegetation composition in this area (Prikryl, 1990: 12). In addition to the

archaeological investigations a floral resources study was performed in immediate

vicinity of a major site. Of the 122 plant specimens collected, 97 were identified and 56

believed to have been utilized, for food, medicines, or technologically by the aboriginal

inhabitants of north-central Texas (McCormick et Al., 1975: 77). These resources are

utilized to aid in the interpretation of the botanical data retrieved from archaeological

sites and their site catchment areas.

For the purpose of the current research it was decided that vegetation would be the

primary determinant of soil economic value given its strength for interpretation and

correlation with soil types. Archaeological faunal data, although economically valuable,

is not incorporated into the determination of relative economic value of the soils even

though it contributes to the knowledge base of prehistoric choices. In the current

research economic value refers to the potential of a given soil type to provide

vegetation that would be economically valuable to prehistoric populations. The

relationship between certain soil types and specific plant types is an assumption that

has been empirically verified by soil scientists and can be tested by future research.

131

The lack of archaeological fauna in relation to all of the sample soil types and soil

settings would have created an inconsistency in the representation of economic values.

The relationship of archaeological fauna to soil type is also less predictable although

future research could test relationships between soil type and archaeological fauna to

try to predict where comparable fauna and their corresponding prehistoric populations

might be located. Faunal data are incorporated into the “overall” economic value and

soils portion of the analysis by mapping the potential for specific wildlife types, in

relation to their habitat, associated with soil types. During the final synthesis the

recoded soils categories, including potential for wildlife habitat, are combined to

determine the overall economic value. However, the correlation between soil type and

economically valuable vegetation is the strongest and provides the highest resolution.

Mapping soils data across a region is one of the most detailed views available of the

environmental character of a particular landscape. Vegetation data, correlated with

soils, sheds light on the potential of the particular types of soils. When the vegetation

is interpreted into use-categories, as related to human populations, the soils provide

culturally significant information. Economic values were determined by evaluating the

botanical, ethnographic and archaeological literature to determine if the sample

vegetation, associated with soil types, was edible, medicinal or supplemental. Edible

vegetation is defined by whether or not the vegetation could be ingested, if it has

served as a food resource for prehistoric or modern populations and the likelihood that

it would be utilized for food based on the relative caloric value. Medicinal vegetation

was defined by the known use of particular plants to be used to cure internal ailments

132

or topical wounds. Many of these plants are highly toxic and require a certain degree of

preparation. Therefore, it was decided not to factor in the toxicity of plants as a

negative impact on whether or not the plant was used by prehistoric populations.

Supplemental vegetation primarily includes wood for fires or construction and plants

that would have been utilized for dyes, basket making, or various types of construction.

Codes are assigned to a sample size of 242 vegetation species collected from

associations made in the soil surveys and archaeological literature. Based on the

edibility, medicinal value, and supplemental potential of the plant the relative values are

translated into economic values per category. The resulting scores are arbitrarily

calculated by assigning a percentage to the applicable category. For example, if a plant

has both medicinal and edible properties each category is given a value of 50 percent.

If a plant is primarily edible with limited medicinal value edibility would be given a value

of 90% and medicinal properties a value of 10 percent. Each percentage is determined

based on the information provided in the descriptive literature. Often a several

resources were evaluated to determine the edible, medicinal and supplemental

properties of the plant in order to take into consideration the likelihood that these

particular plants would have actually been utilized by prehistoric populations. The

resulting scores are provided in the appendices.

Plants are associated with the appropriate soil setting and resulting economic values

are related to soil series. Therefore the 242 vegetation species in the sample are

consolidated into 22 soil settings. The general procedure followed to assign economic

meaning to the landscape is summarized by a conceptual diagram (figure 26). The

133

economic values of the soil settings are linked to the soil series and are proportionally

evaluated within site catchment areas. Ultimately, soil series are consolidated into the

11 great groups that are represented in the site catchment areas so that they may be

summarized. Consolidation of the appropriate soils series provides a relative economic

value for the great groups within each catchment. These values are summed and

assigned to the catchment area to provide an overall economic value. Each catchment

area is also assigned three other thematic economic values (medicinal, edibility,

supplemental) by establishing the same data links. Proportional distribution of the soil

series within the catchment areas is constant. What changes are the economic values

associated with the soil series. This is how the relative edibility, medicinal,

supplemental and economic values are measured in relation to each site catchment.

Economic values and potentials for specific wildlife and vegetation habitat that are

assigned to soils help interpret resource concentration patterns. It has been posited

that sites are primarily located near a major source of food. SCA can shed light on the

estimated economic value of such settings, the kinds of food or environmental

resources that were accessible in the immediate vicinity of site locations and answer

questions related to specialization, diversity and optimization. Furthermore, resource

availability in correlation with site location can shed light on the theories regarding the

use of this region for seasonal migration as suggested by McCormick. Skinner and Baird

hypothesize specialized ecological approaches in the northern reaches of the Elm Fork

correlated with high yield mast crops in the uplands and central locations of the Cross

Timbers. McCormick (1975) noted that sites along the southern reaches of the Elm Fork

134

GREAT GROUP ARGIUSTOLLS CALCIUSTOLLS HAPLUDERTS HAPLUSTALFS HAPLUSTEPTS HAPLUSTERTS HAPLUSTOLLS PALEUSTALFS PALEUSTOLLS RHODUSTALFS USTIFLUVENTS WATER

Site Catchment

Area 1 km2

Soil Series

SOIL SETTING BLACKLAND CLAY LOAM CLAYEY BOTTOMLAND CLAYPAN PRAIRIE DEEP REDLAND DEEP SAND ERODED BLACKLAND LOAMY BOTTOMLAND LOAMY PRAIRIE LOAMY SAND LOW STONY HILL REDLAND ROCKY HILL SANDSTONE SANDY LOAM SANDY SHALLOW CLAY SHALLOW STEEP ADOBE TIGHT SANDY LOAM WATER

SOILS POTENTIAL FOR VEGETATION

POTENTIAL FOR WILDLIFE

ECONOMIC VALUE EDIBILITY

MEDICINAL

SUBSIDIARY

VEGETATIONSOIL SERIES

Figure 26. Procedure to link economic value to site catchment.

are primarily located along the upland edge where maximized resource zones are

accessible. Riparian fauna are associated with archaeological remains along the Elm

Fork indicating concentrations in riverine settings. Sites located along the East Fork

are unquestionably situated in the Blackland Prairie. However, West Fork sites have not

been previously evaluated in prior investigations in relation to their respective

135

physiographic settings. Observation of site patterning along the Elm Fork in correlation

with these settings, on a larger scale, sheds light on regional behavior. Given that

information about these sites is primarily derived from state files there are many

unanswered questions. SCA facilitates the evaluation of site placement within particular

physiographic settings and comparisons of these locations according to occupation

frequency, density and change through time.

Economic values are represented in a series of analytical maps by recoding

attributes associated with the pre-existing soils data layers. These values are

associated with archaeological sites with respect to time period, occupation frequency

and site contents. Then they are statistically evaluated to test a series of hypotheses.

Site Catchment Soils Diversity

Soil series associated with sites provide the best resolution in terms of variation

within a catchment area. Each catchment area contains a square kilometer of soil

series. Each great group contains proportional distribution of soil series relative to the

location of the site. Therefore, the diversity is more accurately expressed by the soil

series. In order to gain a deeper understanding of the diversity within site catchments

each catchment is evaluated with regard to the proportional representation of each soil

series within each site catchment to calculate diversity. Environmental diversity indices

for site catchments measure the potential of an area for the presence of archaeological

sites based on the premise that site location is associated with increased diversity

(Tiffany and Abbott, 1982). For the current research an environmental diversity index

is developed that differs from previous studies in that it is based on the soil proportions

136

of soil types within site catchment boundaries. Given that all of the sites are in the

vicinity of water and that there is a strong correlation between geology, soils and

vegetation a different technique was chosen. This technique is referred to as the

Shannon-Weiner Diversity Index (Colinvaux, 1993: 316),

H1=-Σ pilog2pi,

where pi= relative abundance of soil type i to total soil, is employed to evaluate the

relative diversity of each catchment. By creating an environmental diversity index

based on proportional soil types within the site catchments the current research seeks

to evaluate if there is a correlation between environmental diversity and prehistoric

preferences for selection of site location in north-central Texas.

Site catchment diversity is evaluated in relation to occupation frequency and change

through time in order to determine how variable diversity is at sites that contain

evidence of single versus multiple occupations and how diversity changes through time

in relation to site location. The diversity values in relation to these factors are analyzed

with an ANOVA statistical test.

Geological Variables

Geological formations provide the parent material and foundation for the range of

soils that can develop in a certain location. Given the minor differences in the climatic

regime of this region the majority of variation is dependent on the character of the

bedrock lithology. The geological structures (table 4) that are associated with

archaeological site catchments in the region provide a means to view similarities and

differences in more generalized site settings. This component influences the

137

character of the physiographic setting and is evaluated with respect to change

through time and occupation frequency in a discrete statistical analysis. The West Fork

and Elm Fork of the Trinity are generally associated with similar geological sequences.

Geological formations associated with the East Fork are notably different in character as

listed in table 4. While there is a strong correlation between soils and geological

settings geology is also culturally significant in that it serves a source for lithic

procurement. In addition, it provides a means to cross-check the results of the soils

analysis and to explain the particular variation of correlated soil types. Statistically, the

geology is tested in a different manner than the soils with a Chi-Square analysis rather

than with an ANOVA. As such, geology is evaluated independently of soils even though

they are inter-related.

WEST FORK AND ELM FORK EAST FORKALLUVIUM ALLUVIUM ANTLERS SAND AUSTIN GROUP BOKCHITO FORMATION (UNDIVIDED) FLUVATILE TERRACE DEPOSITS EAGLE FORD FORMATION MARLBROOK MARL FLUVATILE TERRACE DEPOSITS OZAN FORMATION GLEN ROSE LIMESTONE PECAN GAP CHALK GOODLAND LIMESTONE/WALNUT CLAY (UNDIVIDED) WOLFE CITY SAND GRAYSON MARL AND MAIN STREET LIMESTONE KIAMICHI FORMATION PALUXY SAND PAWPAW FORMATION SURFICIAL DEPOSITS (UNDIVIDED) TWIN MOUNTAINS FORMATION WILLOW POINT FORMATION WOODBINE FORMATION Table 4. Geological formations associated with site catchment areas in study area.

138

Physiographic Variables

Physiographic variables that are evaluated in relation to archaeological sites include

the Western Cross Timbers, Grand Prairie/Fort Worth Prairie, Eastern Cross Timbers,

Blackland Prairie, riparian corridors, ephemeral riverine settings and ecotonal settings.

The “Buffer Wizard” in ArcMap is utilized to generate transitional zones around water

resources and physiographic boundaries. Ecotones are generated in ArcMap by creating

2 km buffers around physiographic boundaries. Riparian corridors are generated in

ArcMap by creating 2km buffers around major and minor tributaries in the streams

data. Ephemeral streams are also included but relative to their influence on the

landscape and intermittent character 1km buffers are constructed around these areas.

In the case of the current research physiographic settings include hydrological

resources because these areas are considered to be unique microenvironments.

Physiographic settings are often referred to in the literature in relation to site

location. While this is informative the interpretive value of physiographic settings can

be enhanced with data regarding economic value, faunal assemblage, vegetation

habitat, wildlife habitat inferred from the soils which are related to these physiographic

settings. In addition, larger patterns that emerge in relation to site function, occupation

frequency and change through time can shed light on prehistoric settlement pattern

behavior. A variety of investigators report that grassland, woodland, riverine and

ecotonal settings were utilized by prehistoric populations. Hypothesis testing sheds

light on more specific patterns that emerge in relation to these settings. It is likely that

139

given the various climatic conditions, availability of seasonal resources, and socio-

cultural conditions that certain areas were occupied more intensively than others.

If sites are situated primarily within grassland settings prehistoric populations would

have access to a high variety of grasses, legumes, grains, seed crops, rangeland and

openland wildlife. In addition, site selection along major tributaries would ensure the

availability of supplemental economic resources such as wood for cooking. If the sites

are situated primarily within woodland and riverine settings they would have access to a

high variety of wild herbaceous plants, fish, wetland wildlife and various smaller fauna.

In addition, woodland settings provide a greater abundance of supplemental material

such as wood. Woodland settings also provide access to an increased supply of nuts

and berries. It is expected that the potential for various types of vegetation within

these areas reflects an economically advantageous positioning. Given that both

grassland and woodland settings have a riverine component other factors must be

correlated.

If the pattern detected in the Lewisville report (Ferring and Yates, 1998) that

“prehistoric populations exploited whatever was available to them within well-defined

territories” is true then statistical analyses should reveal that a random effect is

operating. However there may be a pattern that reveals differential utilization of the

landscape. If clearly defined territories exist it is possible to infer the degree of social

influence on these boundaries in relation to site distribution and economic values. If

these populations were primarily effected by socially defined territories economic values

associated with site catchments should reveal that site selection was random in these

140

locations. Further, it is possible to evaluate the environmental variables contained by

site catchments and consider the possibility of environmentally defined territories.

Sites reported in the Lewisville report (Ferring and Yates, 1998) are situated

primarily in the ecotone between the Eastern Cross Timbers and the Blackland Prairie.

Sites in the Ray Roberts area are reported to be centrally concentrated in the Eastern

Cross Timbers. It has been suggested that sites are situated to equally utilize forest,

riparian and prairie environments. If it is true that these environments were equally

utilized then all of these types of soils should be within site catchments. Hypothesis

testing provides a means to determine if similar sites are situated in similar settings.

SCA can help determine what type of site is located in particular settings or where

equivalent sites are located. Sites should reflect the setting through faunal remains

and contents preserved by the archaeological record. Further, it is expected that

economic values of vegetation and wildlife within catchment areas support conclusions.

Physiographic settings are a dynamic variable that interacts with other physical

environmental factors. As such this generalized setting must be evaluated from a

variety of micro-environmental perspectives.

Physiographic settings reflect the inter-relationships between all of the components

that have been forming the modern landscape for millennia. A wealth of information is

related to the available faunal and floral communities contained by these settings. This

particular variable is tested for significance between site catchment areas with a Chi-

Square statistical test. In this case, the goal is to find out if certain physiographic

locations were preferred over others.

141

Climatic Variables

Climatic data available for past environments is sparse but correlations have been

made based on palynological data collected from peripheral locales. Pollen types are

selected from the Ferndale Bog sample only if they are represented in the vegetation

sample utilized for this research. This vegetation is correlated with soil series and

therefore can be depicted on the map. Vegetation and pollen being both associated

with “Plant Family” provided a means to link the results from the Ferndale Bog sample

data (figure 27) to the digital soils data layers and the correlated economic values

associated with site catchment areas. Prior to linking these data to the soil settings

they were calculated to take into consideration the relative representation of these plant

types within site catchment areas. In order to preserve the raw economic values the

pollen data is added to the original values associated with the same settings and

adjusted to reflect the change through time by building economic values for each time

period. These values are depicted a series of maps and correlated with sites in order to

evaluate if sites are situated in areas that are expected to be relatively abundant during

certain climatic intervals. The raw data values and the series of maps that depict these

associations per time period are included in the appendix.

Dawson and Sullivan propose that the character of the environment where

prehistoric sites are located is such that it was necessary for prehistoric groups to

remain small and scattered in order to maintain balance with minimum cycles

associated with climatic and seasonal fluctuations. This type of theory can be tested by

evaluating the correlations between occupation density, economic values and climatic

142

data. Site selection as related to topographic setting can be further enhanced by

incorporating climatic data. For instance, it is suggested that the terrace and upland

settings represent site selection and occupation during wetter climatic episodes. This

inference is tested by relating site locations to the fluctuations of pollen abundance data

based on the Ferndale bog sample. Hypotheses testing can determine if site locations

as they shift and change through time corresponds with climatic conditions.

SCA can reveal if site positioning is a deliberate selection that met the needs of

these populations given the character of the landscape combined with climatic data. If

so, SCA can help ascertain if there is any additional information that facilitates increased

understanding of these choices. If not, such studies can generate evidence to provide

alternative theories.

0% 20% 40% 60% 80% 100%

ACERACEAE

ASTERACEAE

CYPERACEAE

FABACEAE

POACEAE

SALICACEAE

TYPHA

ULMACEAE

EA (WET) MA (DRY) LA (WET) LPI (DRY) LPII (WET)

Figure 27. Relative Abundance of Plant Family in Study Area.

143

Archaeological Variables

Archaeological site coordinates were collected from TARL records, converted to

points in ArcCatalog and projected with ArcToolbox. Site catchment boundaries are

established in ArcMap with the “Buffer Wizard”. Although it is possible to construct time

contours with GIS techniques, the character of the landscape in the study area did not

provide a return on investment equal to the time it took to create the contours.

Therefore, the site catchment area is established for the 150 sites in the study area by

performing a buffering technique to construct an arbitrary 1-kilometer radius around

each archaeological site in the study area. Site catchment areas are constructed for 50

sites that contain diagnostic material and 100 surface sites that do not contain time

diagnostic material. The diagnostic sites are valuable in that they provide clues as to

the specific time period and duration of occupation. The non-diagnostic sites are

valuable with respect to providing evidence of prehistoric activities in relation to

location. When these sites are associated with the specific characteristics of the site

catchment areas they provide fruitful behavioral information in relation to the time

diagnostic sites. One could hypothesize that the sites containing lithic scatter are

ephemeral hunting, gathering, eating or resting spots. Whatever their purpose might

be they provide valuable comparative information in the fine scale analysis of

vegetation and soils data related to site catchment areas. Proximity to time diagnostic

sites also provides an informative view of the possibilities of inter-relationships.

Archaeological site data is categorized into groups to test for statistical significance

between site contents and catchments with respect to environmental variables,

144

occupation frequency and change through time. Activity groups include: tool

manufacture, food preparation, hunting and lithic scatter. Site files are evaluated to

infer activities. Lithic manufacturing activities are inferred if the artifact assemblage at

a site contains cores, hammerstones, performs, bifaces, debitage or a variety therein.

Evidence for food preparation includes tools, groundstone, ceramics, animal bones and

fire cracked rock. Any variety of these items is sufficient for the site to be listed as

including food preparatory activities. Hunting activities are inferred by the presence of

diagnostic projectile points. A separate category of lithic scatter is created to account

for sites where no functional diagnostic information is available. There are enough

cases where various densities of isolated lithic scatter are reported to prompt the

inclusion of this category. An increase in the number of activities represented by tool

types at a site indicates an increase in artifact diversity. It is a basic premise of SCA

that site function and site location are correlated and that inferences can be made

about function from knowledge of location (Roper, 1979:121). Based on previous

studies it is possible to form expectations regarding site function. For instance, sites

that contain a substantial amount of debris have often been classified as “home base”

sites, whether they were occupied year round or only during certain seasons. Previous

investigations report specialized activity sites such as “manufacturing stations” have

been located in the Cross Timbers. Hayes references the archaeology and suggests

that sites on the floodplain represent characteristic mussel procurement sites

referenced by Skinner. He does not specifically reference Skinner’s work but makes an

association between the faunal and artifactual remains, possible activities and

145

topographic settings. Hayes notes that terrace sites consist mainly of lithic scatter and

upland sites are found in areas where lithic raw materials may have been procured.

During the Middle Archaic projectile point size and diversity increases while mobility

generally decreases. It is expected that this type of phenomena is reflected in site

function and location. These types of relationships are investigated and enhanced by

environmental information. Hypothesis testing evaluates site types in other localities to

determine if they are concentrated or randomly distributed. McCormick et al. (1975)

reference similarity of activity at site locations regardless of time period. Factoring in

change through time will shed light on these types of temporal considerations. The

current research correlates site contents with the catchment to determine site function.

Site activities are initially inferred based on the character of the relative artifact and

faunal assemblages. This is intended to provide a simplified means to relate the site

contents to the site catchment. After hypotheses testing site function is verified or

revised.

Occupation density includes ephemeral, moderate and intensive categories. These

categories are formulated on the basis of the number of artifacts reported at a site.

Ephemeral occupation is defined by sites that contain one diagnostic artifact or limited

lithic scatter. Moderate occupation is defined by sites that contain more than one but

less than ten diagnostic artifacts and/or moderate lithic scatter. Intensive occupation is

determined based on more than ten diagnostic artifacts which are often found in

combination with more intensive lithic scatter and a variety of tool types. Previous

reports often reference the density of archaeological content at a site. A model based

146

on correlations with migratory patterns of bison rests on the assumption that these

prehistoric populations were following the herds. Sites related to this pattern are

expected to be more ephemeral in character. SCA facilitates testing for the presence of

temporary camps in relation to seasonal resources and if these sites are fewer in

number than during the Archaic as suggested by McCormick et al. (1975). It is

anticipated that during the Late Prehistoric time period temporary hunting camps will be

found in association with more sedentary base camps due to the opportunities that bow

and arrow technology offers. McCormick also posits that the zone of movement was

along the prairie-cross Timbers boundaries where sites are considered to be “more

permanent”. Sites reported to have the largest collections of Late Prehistoric I artifacts

are located on the Cross Timbers-prairie boundary. Ephemeral sites in some areas are

related to seasonal occupation that focused on harvesting the mast crop resources in

the region. Hypotheses regarding site function and occupation density can test for the

presence of permanent village, temporary campsites and activity-specific sites in the

study area. Occupation density is more meaningful when associated with

environmental attributes present at site locations.

Occupation frequency refers to how many times a site was occupied based on the

number of time periods represented at the site by diagnostic tools or radiocarbon dates.

The categories for occupation frequency include single, dual, tri and multi. Some

investigators have suggested that north-central Texas is a “marginal” setting and others

that the grassland, woodland and riverine settings are “highly opportune”. In order to

evaluate this facet economic value is assigned to soils to identify areas that provide a

147

more marginal or opportune setting. It is possible that during certain time periods the

region provided a marginal setting while during other time periods it provided an

opportune setting. Site density provides a means to evaluate where prehistoric

populations were focusing their efforts during these time periods. If certain locations

provided resources with a relatively higher economic value it is likely that these areas

contain evidence of multiple occupations. Economic values and site density in relation

to change through time provide the basis to test hypotheses that have been formulated

to answer these questions.

Younger surface sites often contain cultural materials of older time periods. This

creates a complex situation that makes it difficult to ascertain if these sites were

occupied by cultures from different time periods or if they represent one culture that

was reusing materials left behind by previous occupants of the area. To add to the

complexity it is unknown if these points were being recycled on location or if they were

collected from other locations and brought back to a central location. It is posited that

identical projectile point types were made by the same individuals or those who shared

technological knowledge. Therefore, it follows that these same individuals would have

also shared knowledge regarding effective plant utilization, hunting techniques and

paired with information regarding suitable locales to exercise these subsistence

strategies. As such it is possible that site catchments that contain the same projectile

point types are either similar in character suggesting that these individuals preferred

certain settings, or very different in character indicating resource concentration in

association with specialized activities and in either case exhibiting a relatively high or

148

low diversity. Hypotheses designed to test for significance evaluate site catchments in

relation to projectile point type.

Established relationships between projectile points and time periods provide the

temporal foundation for the current research. These data are the primary basis for the

evaluation of change through time. Given the types of projectiles found in north-central

Texas research hypotheses look for correlations between point types and environmental

features found in catchments. It is expected that similar point types would be found in

similar settings. If these point types are contemporaneous it is expected that the

people who utilized these tools not only shared technological knowledge but

environmental knowledge. As such it is expected that the prehistoric populations that

utilized these particular point types employed similar subsistence strategies. Given that

the environment would have offered a range of similar optimal opportunities during the

associated time period it is expected that environmental characteristics in association

with similar point types will reflect these opportunities. However, it is also possible that

given seasonal variation strategies will reflect the necessary adaptations. To evaluate

this facet a hypothesis is tested for statistically significant differences between site

catchments in relation to projectile point types. The sites are evaluated to determine

the range of settings that were being utilized during these time periods. Tool kits will

also be evaluated with respect to composition and occupation density to evaluate

activities.

Prikryl (1990) reports that there is a strong cultural continuity between the Late

Archaic and the Late Prehistoric at sites situated on the Elm Fork of the Trinity. Of the

149

36 sites containing Late Prehistoric I artifacts, 34 also contain Late Archaic tools.

Seventeen of the 34 sites were first occupied during the Late Archaic. Similar to the

Late Archaic, 35% of the Late Prehistoric I assemblages are recovered from sites whose

closest water source is the Elm Fork, while the other 65% are recovered from sites

whose closest water source is a tributary stream. Late Archaic settlement patterns

indicate increasing evidence for the reuse of camp-sites and a trend toward seasonal

aggregation of populations when large quantities of food were locally available (Lynott,

1981: 105). Patterns indicate that certain locations were preferred over time (Skinner,

1981) but the nature of these locations is not clear. SCA provides a means to define

these preferences and determine if sites contain evidence of repeated occupation

concentrated in one particular physiographic setting or multiple physiographic settings.

Hypothesis testing identifies favored environments and preferred settings and is

expected to reveal patterns that explain occupation frequency.

Change through time is determined based on the time periods represented by

diagnostic point types recovered from the site. The time periods that are evaluated

during the course of this research include Early (8,500-6,000 BP), Middle (6,000-3,500

BP) and Late Archaic (3,500-1,250 BP), Late Prehistoric I (1,250-750) and Late

Prehistoric II (750-350 BP). Sites are assigned to these time periods based on a variety

of site files, contract reports and academic publications. Evaluating change through

time in relation to site location provides a dynamic view of the interplay of these

variables that respond to climatic and socio-cultural conditions. As such, change

through time is an important facet that can address theories related to site positioning.

150

Correlation of site function and density with site catchment characteristics and change

through time can test theories that have been proposed regarding prehistoric

settlement patterns.

Topographic settings take on enhanced meaning when associated with change

through time. Dawson and Sullivan propose that Archaic sites in the East Fork situated

on ridges and Late Prehistoric sites are situated primarily in bottomland areas. If such

theories are reliable hypothesis testing should provide support or reveal differential

patterning that may be related to other factors.

Physiographic settings are expected to change in response to climatic conditions and

thus exhibit change through time. McCormick et al. (1975) claim that temporary

campsites occupy similar ecological niches and that this remains the case in both

Archaic and “Neo-American” sites. McCormick et al. conclude that Archaic seasonal

camps are located in eco-tonal settings and secondary activity-specific sites situated

with respect to “specific outcroppings, plants and water”. If Archaic camps are

primarily located in eco-tones SCA can help determine the “specific outcroppings,

plants, and water” that they are referring to. Hypothesis testing with respect to change

through time can also test whether or not Late Prehistoric populations follow similar

patterns as McCormick et al. (1975) suggested.

During the Early Archaic, cultural responses to new ecological systems include a

wider variety of projectile points, more variable artifact assemblages, decreased

mobility, seasonal scheduling, and changes in food processing. Subsistence patterns

involve a wider range of plant collection. During the Middle Archaic evidence suggests

151

that a wide variety of vegetation continued to be utilized paired with an increase in

openland economies. Although the importance of hunting appears to be generally

reduced throughout the Late Archaic evidence suggests the continuation of hunting in

the upland prairie. Utilization of a mix of prairie, forest and riparian species is also

indicated during the Late Archaic. Hypothesis testing with respect to site catchment

diversity, economic values and soil potential for wildlife and vegetation should reveal

such patterns or provide alternative perspectives. The added dimension of change

through time delineates types and varieties of resources utilized by prehistoric

populations of different time periods. SCA facilitates evaluation of site placement within

their respective physiographic settings and provides a means to compare these

locations to site content, occupation frequency and density.

A detailed analysis of 100 of the sites, along the Elm Fork and the West Fork of the

Trinity, is performed to evaluate the pattern of site distribution, land use, and possible

function, based on the assemblages and catchments of each site. The purpose of this

portion of the study is to analyze the catchments of diagnostic and non-diagnostic sites

for comparative analysis. The 45 time-diagnostic sites provide a means to observe site

characteristics with respect to change through time and occupation frequency. Since

the East Fork provides a clearly significantly different environmental setting than the

Elm and West Fork the 50 site catchments from this area are included in comparative

discussion but not in statistical analysis. Mapping the economic value of the entire

landscape reveals the overall distribution of valuable resources as they relate to site

location. This can help explain why certain areas were selected over other areas or

152

provide a means to predict likely site locations. Once the intra-site settlement patterns

are evaluated these data can be compared to non-surveyed areas of the Upper Trinity

drainage basin in order to ascertain the kinds of environments associated with the

various prehistoric utilizations of the basin.

The character of investigations in the region introduces a sample bias that

influences the results of the current research. The Elm Fork bridges Ray Roberts Lake

and Lake Lewisville. Professional contract work associated with the construction of

these reservoirs has led to the discovery of a very dense distribution of sites along this

corridor. Academic work conducted at the University of North Texas and Southern

Methodist University has also produced a number of reports along the Elm Fork. Site

distribution in the West Fork portion of the study area is very limited. This is due to the

fact that most of the sites were discovered by amateur archaeologists or land owners.

No formal professional investigations and no contract work, related to reservoir

construction, have been conducted in this portion of the study area. East Fork sites

have been discovered by limited excavation efforts related to academic research. Even

though distribution currently reflects the intensity of archaeological contract work, it is

clear that the Elm Fork found favor with prehistoric populations. Future professional

research should be conducted along the West and East Fork to further explore the

benefits offered in these areas and to substantiate or refute proposed settlement

models. Regardless of sample bias, the current sample provides valuable information

from which to form future hypotheses and conduct research on the settlement patterns

of prehistoric populations in north-central Texas.

153

Site Catchment Analysis of Prehistoric Sites in North-Central Texas

Geographic Information Systems (GIS) permits archaeologists to conduct

sophisticated spatial analyses by providing an extremely useful methodological structure

through which quantitative approaches to environmental attributes and cultural remains

can be spatially addressed. SCA is performed with ESRI ArcView 8.3 ArcMap and SPLUS

2000 statistical software. GIS techniques (figure 28) are an active component,

throughout the entire process of SCA, from data collection to hypothesis evaluation.

GIS provides an ideal set of tools to facilitate the process of performing SCA. The ability

of a GIS to extract environmental information and perform spatial analysis has led to its

utility in SCA. GIS software plays a central role in data collection, input, management

and analysis within site catchments.

Essentially the GIS component provides an organizational mechanism (figure 29) to

manage large data sets from a variety of sources, generate new data sets and integrate

these data to view the dynamic relationships that help explain prehistoric human

behavior. GIS software provides tools to facilitate the construction of databases and a

means to link the archaeological record with the environment, on a regional scale. It is

with these tools SCA is performed and with ecological, anthropological and geographical

theories that settlement patterns are interpreted.

154

DATA

COLLECTION

DATA INPUT

DATA

MANAGEMENT

DATA

ANALYSIS

RESEARCH

HYPOTHESES

OBJECTIVES

MAPPED

INFORMATION

DATA OUTPUT

DECISION

MAKING

ANALOG

DIGITAL

IMPORT

GENERATION

SPATIAL

ATTRIBUTE

SPATIAL

ATTRIBUTE

REPORTS, FILES, MAPS

GIS DATA LAYERS

DATA CONVERSION

RECODING

BUFFERED SELECTION

DATA EXTRACTION

OVERLAY ANALYSIS

POST-PROCESSING

Figure 28. Diagram of GIS data processing design in SCA.

155

GIS Archaeological Sites Archaeological Data

Spatial Locations Economic Data

Statistical and Spatial Analyses

Environmental DataSite Catchment (1km)

Cultural Reconstruction

Cultural Ecology

Archival Information

Archaeological Preservation

Environmental Reconstruction

Model of Regional Prehistoric Settlement

Pattern Behavior

Figure 29. GIS in relation to components of the settlement model.

156

In general, the SCA, which is a synthesis of archaeological, environmental and

ethnographic data, includes the following procedures:

1. Soil survey soil series level data are evaluated with respect to vegetation.

2. Vegetation, associated with specific soil types in the soil surveys are identified by common name, family, genus and species.

3. Research is conducted to determine the relative values, per plant species, of

edibility, medicinal value and supplemental value.

4. Values are attached to soil series level data in ArcMap GIS interface to the digital soils data layer.

5. Soil series are recoded with economic values, vegetation and wildlife potentials

using ArcMap “joining” operations to relate to site distribution in light of chronology, site frequency and artifact density.

6. Archaeological sites are projected with ArcCatalog and ArcToolbox and then

imported into the ArcMap GIS interface where 1 km catchment areas are constructed around the sites with ArcMap buffering techniques.

7. Soil types within catchments are proportionately extracted, recorded and

interpreted into measures of diversity, plant and wildlife potential and economic worth with ArcMap selection and extraction techniques.

8. Overlay analysis is performed using ArcMap to provide a visual representation of

the intersecting data layers that form the environment in association with sites and facilitate evaluation of spatial patterning.

9. Observable topographical, geological, hydrological, physiographic and ecotonal

resources are related to sites chronologically, with respect to occupation frequency and assemblage composition, by selection and extraction ArcMap techniques to compile contingency tables for chi-square analyses.

10. Descriptive and multivariate statistical tests are performed with SPLUS 2000

software to objectively compare the economic values, wildlife potentials, vegetation potentials, topographic settings, great groups, geological and physiographic settings within site catchment areas to look for patterns in site occupation, composition and change through time.

157

Results generated from the SCA are, in part, presented in a series of maps. A series of

thematic databases, charts, histograms, and graphs summarize the proportional

distribution of soil types, geological settings, topographic settings, physiographic

settings, economic potentials and catchment soil diversity. Site catchments are

summarized with descriptive and multivariate statistical techniques to describe, quantify

and compare the resource potentials of site territories. Based on the prior research a

series of hypotheses are tested and a model of prehistoric settlement patterns is

formulated.

Hypotheses Testing

In order to understand prehistoric human socio-economic settlement patterns in

north-central Texas this thesis specifically aims to formulate a model that can help

explain why prehistoric populations selected certain locations for single or repeated

occupation between the Early Archaic (8,500 BP) and Late Prehistoric (500 BP).

Explaining site selection is the driving research problem because location influences the

range of available economic opportunities and cultural adaptations. In order to address

this challenge a series of hypotheses have been formulated. These hypotheses are

designed around two general themes. The first theme is descriptive while the second

theme is interpretive.

Description oriented hypotheses focus on the descriptive character of the site

catchments to define the features related to site re-occupation and chronological

change. The first set of hypotheses is formulated to evaluate the physical

characteristics (topographic, physiographic, geological, great group and soil setting) of

158

site catchments in relation to occupation frequency (single, dual, tri, multi) and change

through time (from Early Archaic to Late Prehistoric II). This set of hypotheses is

formulated to provide a means to determine the specific physical characteristics

typically associated with sites that have been variably occupied and to provide a view of

how this changes through time. Evaluation of the relationships between site

catchments and the prehistoric activities that occurred at these locations require

hypotheses designed to examine site catchments in relation to site function (lithic

scatter, tool manufacture, hunting, food preparation) occupation density (ephemeral,

moderate, intensive) and cultural affinities (projectile point types). In order to factor in

the element of time, site function and occupation density are evaluated in relation to

occupation frequency and change through time.

Interpretive hypotheses are designed to test change through time and occupation

frequency in relation to potentialities for prehistoric utilization of certain environmental

resources. Environmental information is translated into a meaningful ecological context

related to human decision-making processes. Essentially, this portion of the analysis is

designed to test relationships between site locations and interpreted soil potentials.

Therefore the data are extended beyond the physical to the possible. Potential for

habitat suitable for certain types of wildlife and vegetation provides general insight on

what types of animals and vegetation were suited to certain soils.

Economic value extends interpretation beyond soils to vegetation linking these data

to possible prehistoric utilizations of plants that have the potential to exist given certain

soil types. Climatic interpretations link soils data to pollen data providing a means to

159

estimate the relative abundance of certain plants during wetter or drier climatic

intervals that have been documented in correlation with past patterns. Wildlife

(openland, rangeland, wetland); vegetation (grains and seed crops, grasses and

legumes, wild herbaceous plants); economic value (edibility, medicinal, supplemental);

and climatic interpretations of soils data are evaluated in relation to site catchment

characteristics, site contents, occupation frequency and change through time.

Relationships between the diversity within site catchment areas and the variability of

archaeological assemblages are examined. Variability between site catchments relative

to occupation frequency and change through time is evaluated. Defining the character

of the site catchment areas to determine diversity within and variability between these

areas helps explain why sites were selected repeatedly or ephemerally. Interpreting

the relative economic value of these variables provides a means to evaluate prehistoric

behavioral choices and adaptations as they change or stay the same through time. A

multi-scale investigation into landscape diversity within and variability between site

catchments of archaeological sites is selected to test these hypotheses. As such, the

SCA requires the ability to superimpose the archaeological record upon the micro-

environmental setting. Collectively, cultural and environmental data are analyzed, with

a unique methodology, to shed light on the character of multi-dimensional expression of

prehistoric decision-making processes.

As a part of the SCA process, environmental modeling is expected to reveal that

archaeological sites are strategically situated in areas that yield protection or in areas

that provide advantageous economic positioning. Correlations are expected between

160

select topographic settings, physiographic settings, soil potential, economic values and

climatic conditions. Given the physiographic character of north-central Texas, it is

expected that site selection strategies should reflect an equal utilization of the ecotones.

In relation to this, increased environmental diversity within the site catchment area is

expected. On the other hand, it is possible that certain areas were utilized for

specialized resource gathering. If this is the case, decreased environmental diversity is

expected. SCA also facilitates the inference of site function based on the economic

potential of the microenvironments that surround each site in conjunction with the

assemblages from them. It is anticipated that decisions made in relation to

environmental structure will provide comparable inferences regarding settlement

patterns to those drawn by other hunter-gatherer settlement studies.

In order to test the hypotheses archaeological and environmental data layers are

evaluated to test for differences between the environmental settings of site locations in

relation to assemblage composition. Occupation densities combined with activity

groups represent cultural behavior. Economic values (medicinal, edible, supplemental);

soil potential (vegetation and wildlife habitat); physiographic settings, topographic

settings, soil settings and geological settings are environmental attributes that are

factored into the statistical analyses. Hypotheses are either accepted or rejected on the

basis of how well the distribution of soils, geologic, hydrologic, topographic,

physiographic and economic variables met expectations in relation to assemblage

composition (lithic scatter, lithic manufacture, hunting, food preparation); occupation

density (ephemeral, moderate, intensive); occupation frequency (single, dual, tri, multi)

161

and change through time (Early Archaic-Late Prehistoric II). Variables and methods

utilized to evaluate each null hypothesis are addressed and summarized in this section.

Catchment diversity values, economic values, soil potentials, soil settings, great

groups, geological, physiographic and topographic data layers are utilized to test the

following hypotheses (table 5). The first predicts differences between site catchments

with respect to visitation frequency. The second evaluates change through time. The

third factors in site function. The fourth compares relationships and occupation density.

Null Hypothesis

There is no statistically significant difference between site catchments in relation to:

1 occupation frequency 2 change through time 3 site function 4 occupation density Table 5. Site catchment hypotheses.

Economic data layers are evaluated to look for relationships that may emerge regarding

specialized or generalized activities. Diversity data layers are evaluated to look for

relationships between catchment diversity and site occupation. Economic and diversity

values are both expected to correspond with increased site occupation frequency. An

ANOVA is conducted to evaluate economic values and diversity values in relation to

occupation frequency. A series of Chi-Square statistical tests are performed to test for

statistical significance of the great groups, soils, geological, topographic and

physiographic settings within site catchments in relation to occupation frequency. If

differences are statistically significant, frequency tables are evaluated to determine

what type of site selection patterns emerge in relation to these factors.

162

Chi-Square statistical tests are performed to test for statistical significance of site

function and occupation density in relation to change through time. If differences are

statistically significant, frequency tables are evaluated to determine what type of site

selection patterns are prevalent. These hypotheses are expected to shed light on the

relationship between site function and occupation density in relation to change through

time and occupation frequency (table 6) without factoring in the environment.

Null Hypothesis There is no statistically significant difference between site function

in relation to: 5 change through time 6 occupation frequency Null Hypothesis There is no statistically significant difference between occupation

density in relation to: 7 change through time 8 occupation frequency

Table 6. Site function hypotheses.

This facet is important because it provides a statistical view of prehistoric occupations

as they change through time. It reveals if inferred prehistoric activities and their

associated densities are similar or different with respect to change through time and in

relation to occupation frequency.

ANOVA is performed on a collection of databases that contain time diagnostic site

catchments and their associated proportional wildlife (openland, wetland, and

rangeland) potentials; vegetation (grains and seed crops, grasses and legumes, wild

herbaceous plants) potentials; and economic values (edible, medicinal, supplemental).

This multi-variate statistical test will evaluate a series of hypotheses (Table 7).

163

Null Hypothesis There is no statistically significant difference between wildlife potentials associated with site catchments in relation to:

9 change through time 10 occupation frequency Null Hypothesis There is no statistically significant difference between

vegetation potentials associated with site catchments in relation to:

7 change through time 8 occupation frequency Null Hypothesis There is no statistically significant difference between economic

values associated with site catchments in relation to: 9 change through time 10 occupation frequency

Table 7. Economic values hypotheses.

Determining the relationship between the character of site catchment areas and the

variability of archaeological assemblages relative to occupation frequency and change

through time provides explanatory mechanisms for prehistoric settlement patterns.

Purposeful behavior is patterned and based on socially transmitted information that

facilitates successful adaptations. If site selection is the result of purposeful human

activity, it should likewise be patterned. Although it is expected that the patterning of

site selection is reflective of the human behavior that produced it, these similarities

must be demonstrated and not assumed. Human actions are processes in time and

space that have structure and can be presented as a network of inter-related behavior.

Inferences can be made based on what is known about the character of the

environments that provided the range of possibilities for prehistoric adaptations. When

site distribution is calibrated with the environmental setting, given a variety of

scenarios, it is possible to build a model of regional prehistoric settlement patterns.

164

Study Area

The area in north-central Texas (figure 30), selected for this research, is defined by

the three major tributaries of the Trinity that intersect three distinctive physiographic

zones. For the most part sites are situated along or near these waterways. On a

regional scale, SCA is performed on 150 surface sites selected from the West Fork, Elm

Fork and East Fork of the Trinity, with projectile point typology as the means to

establish chronology and occupation frequency. Data gathered from these regional

site locations are evaluated using 10 excavated sites from the Elm Fork of the Trinity

that produce a well-documented and well-stratified chronological record. These

excavated sites provide a means of temporal control for the regional model. The time

span of this study ranges from 8500 years before present, during the Early Archaic, to

the Neo-American era, 500 years before the present era. Excavated sites selected for

this research provide a discontinuous record of occupations, in north-central Texas,

from the Early Archaic to Late Prehistoric. Excavated sites are useful for this research

since they provide a detailed account of artifacts, features, fauna and radiocarbon

dated materials. These materials are particularly suited to SCA because the level of

detail facilitates a more complete reconstruction of economic activities in space and

through time. In addition, excavated sites provide empirical field data as a means to

test validity of the extrapolated data derived from digital map layers that form the basis

for the analysis. Selected study sites facilitate a chronological examination of diversity

within site catchment areas and variability between site locales to consider the variety

of factors that could influence site selection. In addition, an opportunity to examine site

165

Figure 30. North-central Texas research study area.

166

catchments for evidence of functional constants that may be associated with site

selection is provided. The regional scale analysis that includes the 150 surface sites

provides a basis from which to infer site type, function, examine trends and evaluate

the environmental diversity and relative economic value on a regional scale. According

to Ferring (1986), archaeologists must continue to develop research methods that

enhance the study of intra-site variability with respect to the larger, multi-site cultural

system. Although, Ferring is referring to artifacts and fauna, the same concept is true

for the environmental setting. This research views the catchment area as an extension

of the archaeological remains. As such the results of the SCA are an important facet of

the regional settlement pattern model formulated for this region.

167

CHAPTER 6

RESULTS

The most analytically informative data produced by conducting the site catchment

analysis (SCA) in north-central Texas is contained by a series of maps, graphs and

tabular data. In this chapter results are evaluated with respect to each variable that is

evaluated either spatially or statistically and contributes to the development of the

regional settlement pattern model. A discussion of the distribution of sites in relation to

the geological, physiographic, topographic and soils settings, as well as, site function

and site activities are followed by a synopsis of the statistical results.

Geological Setting

The geological setting of sites in the study area (figure 31) provides tremendous

utility for environmental reconstruction and behavioral interpretation. Closer observation

of this variable reveals strategic decision-making. Sites located in the West Fork study

area are predominantly situated in sandstone settings with secondary presence in

alluvial settings and tertiary presence in limestone (figure 32-33). Elm Fork sites follow

a similar pattern however shale settings replace limestone settings. East Fork sites are

typically associated with alluvial settings, marl or a combination therein.

In the Elm Fork study area 48% of the Archaic sites are on fluvatile or alluvial

deposits, 41% on sandstone and 9% on shale. Late Prehistoric sites are situated in

water deposited sediments (40%), sandstone (45%) and shale (13%). Hunting sites in

the Elm Fork area are fairly equally represented in both water deposited sediments

(44%) and sandstone settings (37%) with the remaining in shale (9%) and limestone

168

(3%) settings. Lithic manufacture camps are primarily associated with sandstone

(45%) with a secondary presence (34%) in alluvial and fluvatile terrace deposits.

In the West Fork area 56% of the Archaic sites are associated with sandstone and

22% with water deposited sediments. Of the Late Prehistoric sites 25% are found in

association with alluvium or fluvatile terrace deposits and 50% with sandstone. Sites

that contain evidence of lithic manufacture, hunting and even base camps are found in

association with sandstone (75%) with the remaining 25% found on fluvatile terrace

deposits. Only one site was found in relation to limestone and upon closer inspection it

appears to be situated directly on the dividing line between the Goodland Limestone

Formation and the Antlers Sandstone. Sites in the West Fork area are actually a very

good example of sites oriented in an ecotonal setting with a notable precision.

Interestingly, in almost every case cases the sites are on the sandstone side of the

ecotone about a half of a kilometer from the limestone. The alternative scenario is

illustrated by one site in particular situated almost half way between the West Fork and

the Elm Fork. In this case, where Goodland limestone dominates the setting, a small

fluvatile terrace deposit was selected and the site was placed on the edge overlooking a

stream valley. This is an excellent example of how geology can be used to reconstruct

prehistoric behaviors that appear to be influenced by past environments.

In the East Fork portion of the study area Archaic sites tend to be associated with

alluvium (27%) and fluvatile terrace deposits (33%). An opposite pattern occurs during

the Late Prehistoric when sites are primarily located in alluvial deposits (34%) with a

169

Figure 31. Geological Setting of Sites in Study Area.

170

Figure 32. Geological Settings and Time Periods. This figure illustrates the relationships between the number of sites that contain each particular geological deposit. For the sake of clarity in illustration combined with the fact that there is a significant overlap of site selection with respect to change through time it was necessary to consolidate the subdivisions of the particular time periods. Early, Middle and Late Archaic are grouped into “Archaic”. The first and second phase of the Late Prehistoric is combined into the “Late Prehistoric” category. The “Prehistoric” “category consists of sites that do not sites are associated with marl. Although functional differences between the Prehistoric

GEOLOGICAL SETTING: ELM FORK

10

20

30

40

50

60

WATERDEPOSITEDSEDIMENTS

LIMESTONESETTINGS

SHALESETTINGS

CLAY RICHLIMESTONESETTINGS

SANDSTONESETTINGS

SIT

ES (

CO

UN

TS)

ARCHAIC LATE PREHISTORIC PREHISTORIC

GEOLOGICAL SETTING: WEST FORK

10121416182022

WATERDEPOSITEDSEDIMENTS

LIMESTONESETTINGS

SHALESETTINGS

CLAY RICHLIMESTONESETTINGS

SANDSTONESETTINGS

SIT

ES

(CO

UN

TS)

ARCHAIC LATE PREHISTORIC PREHISTORIC

contain diagnostic materials. Geological deposits are also consolidated into their more generalized “Parent” groups.

171

Frc

GEOLOGICAL SETTING: ELM FORK

10

20

30

40

50

60

WATERDEPOSITEDSEDIMENTS

LIMESTONESETTINGS

SHALESETTINGS

CLAY RICHLIMESTONESETTINGS

SANDSTONESETTINGS

SIT

ES (

COU

NTS

)

MULTI-COMPONENT SINGLE-COMPONENT

GEOLOGICAL SETTING: WEST FORK

10121416182022

WATERDEPOSITEDSEDIMENTS

LIMESTONESETTINGS

SHALESETTINGS

CLAY RICHLIMESTONESETTINGS

SANDSTONESETTINGS

SIT

ES (

COU

NTS

)

MULTI-COMPONENT SINGLE-COMPONENT

igure 33. Geological Settings and Occupation Frequency. This figure illustrates the elationship between geological deposits and occupation frequency. Sites are onsolidated into single or multiple occupancy and geology into generalized lithology.

172

secondary presence in fluvatile terrace deposits (19%). In addition, 26% of the Late

Prehistoric sites are associated with marl. Although functional differences between the

sites along the East Fork are not currently clear due to lack of data there is a strong

correlation between the four sites that contain the "Wylie Focus Pits" and the Ozan Marl

formation. This is significant due to the fact that most of the landscape is dominated

by the Austin Chalk to the West and Marlbrook Marl to the East. However, given that

alluvial deposits and the associated major tributaries are carved into the Ozan Marl, this

appears to be a logical choice that provided immediate access to water resources. In

addition, this geological deposit is less resistant than the surrounding geology which

would have facilitated the construction of the characteristic deep ceremonial pits.

Geology in correlation with occupation frequency follows similar patterns as change

through time. This is probably due to the fact that sites that exhibit reoccupation are

occupied during more than one time period and the majority of sites with diagnostic

materials exhibited reoccupation. This is a problematic facet of the current sample. As

such, it is emphasized that the bias in site distribution is influencing the overall trends

but when the sites are observed individually there is clearly a pattern of strategic site

selection with respect to specific geological settings. In this case, sites are typically

situated in water deposited sediments nested within sandstone settings.

Physiographic Setting Physiographic (figure 34) settings of the Elm Fork include three vegetation zones.

East to West they are: Blackland Prairie, Eastern Cross Timbers and Grand Prairie or

Fort Worth Prairie. Most of the Elm Fork of the Trinity is included in the Eastern Cross

173

Timbers. The Elm Fork of the Trinity includes a wide band of the Eastern Cross Timbers

flanked on the west by the Grand Prairie and on the east by the Blackland Prairie.

Regardless of the effects of change through time (figure 35) and occupation frequency

(figure 36) most of the archaeology along the Elm Fork is in the ecotone between the

Cross Timbers and Prairies.

The West Fork of the Trinity (figure 34) is in a transitional zone between the

Western Cross Timbers and the Fort Worth/Grand Prairie. The cultures that occupied

this area lived precisely within the eco-tone between woodlands and prairies even when

change through time (figure 35) and occupation frequency (figure 36) are considered.

Sites situated in these areas have a high degree of flexibility between the prairies and

the woods increasing the success of various subsistence strategies.

The East Fork of the Trinity is situated in the center of the Blackland Prairie. The

only access to woodland species is along riparian corridors. As such sites tend to

aggregate along the main stem of the East Fork (figure 34) rather than along minor

tributaries and ephemeral streams. Bottomland hardwoods also provide access to the

benefits offered by a wooded environment. Physiographically, the options open to

prehistoric populations were limited. However, the variation in the limestone and marl

geology, combined with the riparian settings, provided a fairly diverse environment.

Utilization of the ecotone occurs along both the West Fork and the Elm Fork.

Change through time appears to be more affected by the number of known sites than

due to prehistoric preferences. Ecotonal settings on the fringes of the eastern and

western edges of the Cross Timbers appear to be the most heavily utilized areas along

174

the Elm Fork. Sites are equally distributed between the two prairie/woodland ecotones

with a scatter of centrally located sites in the Cross Timbers.

Figure 34. Physiographic settings in north-central Texas study area.

175

Minor riparian corridors (41%) in both the West and the Elm Forks are utilized most

often. Major River corridors (30%) and ephemeral stream settings (29%) are utilized

fairly equally. Along the East Fork there are no ecotonal areas immediately present

unless the riparian environments are included in this category. East Fork sites are

PHYSIOGRAPHIC SETTING: ELM FORK

1020304050607080

PRAIRIES CROSSTIMBERS

MAJORRIPARIAN

CORRIDORS

MINORRIPARIAN

CORRIDORS

ECOTONE

SITE

(CO

UN

TS)

ARCHAIC LATE PREHISTORIC PREHISTORIC

PHYSIOGRAPHIC SETTING: WEST FORK

10121416

18202224

PRAIRIES CROSSTIMBERS

MAJORRIPARIAN

CORRIDORS

MINORRIPARIAN

CORRIDORS

ECOTONE

SITE

(CO

UN

TS)

ARCHAIC LATE PREHISTORIC PREHISTORIC

Figure 35. Physiographic settings and time periods.

176

PHYSIOGRAPHIC SETTING: ELM FORK

1020

30405060

7080

PRAIRIES CROSSTIMBERS

MAJORRIPARIAN

CORRIDORS

MINORRIPARIAN

CORRIDORS

ECOTONE

SIT

E (

CO

UN

TS)

MULTI-COMPONENT SINGLE-COMPONENT

PHYSIOGRAPHIC SETTING: WEST FORK

1012

1416

1820

2224

PRAIRIES CROSSTIMBERS

MAJORRIPARIAN

CORRIDORS

MINORRIPARIAN

CORRIDORS

ECOTONE

SIT

E (C

OU

NT

S)

MULTI-COMPONENT SINGLE-COMPONENT

Figure 36. Physiographic settings and occupation frequency.

situated primarily within grasslands along major tributaries (26%), minor tributaries

(26%) and ephemeral streams (52%) providing immediate access riverine resources.

Observations from previous investigations that reported grassland, woodland,

riverine and ecotonal utilization are substantiated. The observation that sites are

situated to equally utilize forest, riparian and prairie environments appears to be true in

177

relation to the Elm and West Forks but is not the case along the East Fork. Although

there is inherent ecotonal utilization in the East Fork area by virtue of the riparian

corridors nested within the prairie environment. Given the spatial patterning and

reoccupation of sites through time it is likely that the environment was a more

significant factor then socially defined boundaries. Although, it is also possible, very

much like today, that environmental boundaries coincided with social boundaries. It

appears that these populations strategically placed themselves so they would have

access to a variety of economically valuable resources.

Topographic Setting

Topographic settings of site catchments in the Elm Fork (figure 39) and West Fork

(figure 40) provide a means to compare the orientation of sites from a variety of

perspectives. Archaic and Late Prehistoric sites located on the West Fork are

predominantly situated in terrace (47%) and upland (26%) settings near primary and

secondary water sources. Drainage settings (21%) and floodplain (5%) settings are

represented by 3 single and 2 multi-component sites. Archaic and Late Prehistoric Elm

Fork sites are predominantly located in upland (45%) and terrace (41%) settings. Only

12% are located in the floodplain and 1% in drainages. Archaic East Fork populations

selected alluvial terraces (41%), floodplains (25%), and uplands (33%). During the

Late Prehistoric a slight decrease in terrace settings to 38% is countered by an increase

in floodplains to 30% while upland settings (32%) remain about the same. Fairly

equal distribution of sites in upland and terrace settings in both the West and the Elm

Fork exist regardless of change through time (figure 37). Overlapping of time

178

diagnostic sites creates complexity in this respect. A similar distribution between

topographic settings occurs in relation to occupation frequency that is fairly consistent

between the East, West and Elm Forks of the Trinity (figure 38).

TOPOGRAPHIC SETTING: ELM FORK

10

20

30

40

50

60

70

BOTTOMLANDS WATER(DRAINAGES)

FLOODPLAINS UPLANDS TERRACES

SIT

E (

COU

NT

S)

ARCHAIC LATE PREHISTORIC PREHISTORIC

TOPOGRAPHIC SETTING:WEST FORK

1012141618202224

BOTTOMLANDS WATER(DRAINAGES)

FLOODPLAINS UPLANDS TERRACES

SITE

(C

OU

NT

S)

ARCHAIC LATE PREHISTORIC PREHISTORIC

Figure 37. Topographic Settings and Time Periods.

179

TOPOGRAPHIC SETTING: ELM FORK

1020304050607080

BOTTOMLANDS WATER(DRAINAGES)

FLOODPLAINS UPLANDS TERRACES

SIT

ES

(C

OU

NTS

)

MULTI-COMPONENT SINGLE-COMPONENT

TOPOGRAPHIC SETTING: WEST FORK

1012141618202224

BOTTOMLANDS WATER(DRAINAGES)

FLOODPLAINS UPLANDS TERRACES

SIT

ES

(CO

UN

TS

)

MULTI-COMPONENT SINGLE-COMPONENT

TOPOGRAPHIC SETTING (EAST FORK)

05

1015202530354045

ALLUVIAL TERRACES FLOODPLAINS UPLANDS STREAM TERRACES

SIT

ES

(C

OU

NTS

)

MULTI-COMPONENT SINGLE-COMPONENT

Figure 38. Topographic Settings and Occupation Frequency.

180

Figure 39. Elm fork site catchment elevations, change through time and projectile frequency. This figure is a screen shot of the GIS interface where the sites are evaluated from two perspectives (change through time and frequency of diagnostic points) juxtaposed over top the elevation. The larger circles are meant to show the clusters that are occurring per time period. With a few exceptions, catchments outside of the clusters are non-diagnostic general prehistoric sites.

181

Figure 40. West fork site catchment elevations, change through time and projectile frequency. This figure is a screen shot of the GIS interface where the sites are evaluated from two perspectives (change through time and frequency of diagnostic points) juxtaposed over top the elevation. In the West Fork portion of the study area there is very little diagnostic evidence as reflected by the lack of indication of change through time. Only two sites contain evidence of change through time.

182

Results support the observations made during previous investigations that suggest that

prehistoric populations primarily selected sites located in terraces and upland settings in

ecotonal transitional zones between prairie and woodland settings. Terrace and upland

settings along both the West and Elm Fork of the Trinity indicate reoccupation. East

Fork sites contain evidence of occupation in uplands, floodplains, and terrace settings

near bottomland riparian environments. In this respect the East Fork is very similar to

the West and Elm Forks of the Trinity.

Soils: Great Groups Soil taxonomy provides a means to interpret the potential of soils to produce

vegetation and provide habitat for wildlife. Given the similar geological foundations of

the Elm and West Fork results indicate a fairly equal correlation with Alfisols, Mollisols

and Vertisols. Paleustalfs, Haplusterts, Haplustolls, Haplustalfs, Hapluderts and

Calciustolls. All of these great groups have a common denominator that indicates a

fairly moist environment with dry intervals probably related to seasonality or climatic

fluctuations. Evaluating soil distribution (figure 41) for the West and Elm Forks across

the study area and in the diagnostic catchment areas provides an overview of the

variability. It is the distribution of these soils in relation to sites that determines the

relative accessibility to economically valued resources by prehistoric populations. The

primary difference between the great groups in the study area regardless of change

through time (figure 43) and occupation frequency (figure 44) is in the level of horizon

development and soil order. Otherwise in all portions of the study area there is a

proportional distribution of Alfisols, Mollisols, Vertisols and Inceptisols.

183

Figure 41. Great group distribution and site catchments.

184

East Fork soils (figure 42) are considerably different from the West Fork and Elm Fork.

Pellusterts, Chromuderts and Chromusterts indicate a strong presence of Vertisols with

varying degrees of organic content. The ustic component indicates a fairly moist

setting with dry intervals whereas the udic setting indicates a higher moisture regime

Figure 42. East Fork great group distribution and site catchments.

185

GREAT GROUPS: ELM FORK

1020304050607080

MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS

SIT

ES (

COU

NT

S)

ARCHAIC LATE PREHISTORIC PREHISTORIC

GREAT GROUPS: WEST FORK

10

15

20

25

MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS

ARCHAIC LATE PREHISTORIC PREHISTORIC

GREAT GROUPS: EAST FORK

-5

5

15

25

35

45

VERTISOLS MOLLISOLS ALFISOLS INCEPTISOLS

SIT

ES (

COU

NT

)

ARCHAIC LATE PREHISTORIC PREHISTORIC

Figure 43. Great groups and time periods.

186

year-round. Haplaquolls and Haplustolls represent the typical Mollisols that are

associated with the Austin Chalk. Hapluquolls are associated with a wetter environment

then the Haplustolls probably indicating a closer proximity to the proximal water

resources. Higher organic content is associated with these taxonomic units which is

attributable to vegetation associated with the characteristic prairie setting.

Comparatively East Fork sites provide a significant contrast to the West and Elm Fork

of the Trinity. The East Fork is dominated by Vertisols with a secondary presence of

Mollisols and tertiary presence of Alfisols regardless of change through time (figure 43).

In relation to this it is interesting to note that multi-component sites tend to have a

fairly equal distribution of great groups indicating that these populations were probably

strategically selecting base camp locations in areas that provided access to the highest

diversity of vegetal resources. On the other hand single component (figure 44) site

catchments all contain Vertisols. West Fork single component sites show more variation

in the great groups within their respective catchments whereas the multi-component

sites have more of an equal distribution of great group types with a slightly greater

emphasis of Alfisols. Given that these sites are associated with sandstone settings this

situation is anticipated. In addition, the number of sites situated on the ecotone, along

the Elm Fork, there is a fairly equal distribution of great groups within the site

catchments regardless of occupation frequency (figure 44). Soils provide a situation that

is quite different from the more generalized categories. This variable has much more

overlap than any other due to the level of detail it provides. In all cases site

catchments contain more than one soil series and more than one great group.

187

GREAT GROUPS: ELM FORK

1020304050607080

MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS

SITE

S (C

OU

NTS

)

MULTI-COMPONENT SINGLE-COMPONENT

GREAT GROUPS: WEST FORK

0

2

46

8

10

12

MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS

SIT

ES (

COU

NT

S)

MULTI-COMPONENT SINGLE-COMPONENT

GREAT GROUPS: EAST FORK

0

10

20

30

40

VERTISOLS MOLLISOLS ALFISOLS INCEPTISOLS

SIT

ES (

CO

UN

TS)

MULTI-COMPONENT SINGLE-COMPONENT

Figure 44 Great groups and occupation frequency.

188

Soils: Soil Settings

Soil settings in the Elm and West Fork are fairly similar. Due to the level of detail

provided by the soil surveys the East Fork is not included in this portion of the analysis.

The most dominant groups are clay-rich and loamy settings. Sandy soils for the most

part have a high clay content which has created soils that are loamy in character. There

appears to be an equal representation of each setting along the Elm and West Forks in

relation to change through time (figure 45) and occupation frequency (figure 46).

SOIL SETTING: ELM FORK

1020304050607080

CLAY RICHENVIRONMENTS

SANDYENVIRONMENTS

LOAMYENVIRONMENTS

OTHER

SITE

S (C

OU

NTS

)

ARCHAIC LATE PREHISTORIC PREHISTORIC

SOIL SETTING: WEST FORK

1012141618202224

CLAY RICHENVIRONMENTS

SANDYENVIRONMENTS

LOAMYENVIRONMENTS

OTHER

SITE

S (C

OU

NTS

)

ARCHAIC LATE PREHISTORIC PREHISTORIC

Figure 45. Soil settings and time periods.

189

SOIL SETTING: ELM FORK

10

20

30

40

50

60

70

80

CLAY RICHENVIRONMENTS

SANDYENVIRONMENTS

LOAMYENVIRONMENTS

OTHER

SIT

ES (

COU

NT

S)

MULTI-COMPONENT SINGLE-COMPONENT

SOIL SETTING: WEST FORK

10

12

14

16

18

20

22

24

CLAY RICHENVIRONMENTS

SANDYENVIRONMENTS

LOAMYENVIRONMENTS

OTHER

SIT

ES

(CO

UN

TS

)

MULTI-COMPONENT SINGLE-COMPONENT

Figure 46 Soil settings and occupation frequency. Soil settings from the West and Elm Forks reflect the utilization of eco-tonal settings.

These characteristic settings are essentially the basis for vegetation correlations. As

such, the frequency of sites associated with a given soil setting provides important

information regarding prehistoric preferences of habitat. The general categories

illustrated here are intended to show overall trends. Interpretive views are available

when this data is evaluated in correlation with the types of associated economic values.

190

Soils: Economic Interpretations

Economic interpretations for soils provide enlightening information in relation to site

distribution. While there is not a statistically significant difference, between site

catchments, there are similarities that indicate strategic selection of locations on the

landscape that provide important resources to sustain prehistoric populations. One of

the few statistically significant results between site catchments are associated with soil

settings. This is directly related to the particular economic value of the site catchment

as established by the current research. Mapping of the distributions of economic value

with respect to edibility, medicinal value, supplemental value, wildlife and vegetation

provides visual cues that reveal patterns. These patterns can be viewed in relation to

change through time, occupation frequency, density, projectile point distribution and a

variety of other facets. A series of maps are provided in the appendix (A1-A22) to

illustrate the relationships of the sites to the resources relative to their interpretive

values as site location changes through time. Maps provided in this section illustrate

relationships between soil values and occupation frequency, as well as, site locations

and varying degrees of economically valuable resources. Graphs provided in this section

summarize data extracted from the site catchment areas in relation to occupation

frequency and change through time.

Sites are evaluated in relation to the relative economic values represented within the

catchment areas. Soils data are recoded to illustrate (figure 47) the overall distribution

of edible, medicinal and supplemental vegetation patterns on a larger scale in relation

to the site locations. Darker values indicate a lower potential for these specific types of

191

Figure 47. Economic values and prehistoric components.

192

vegetation whereas the lighter values indicate an abundance of the associated

resources. The lower values are weighting the sample due to the lack of soils data

where reservoirs now exist. Change through time with respect to these variables is

illustrated in the appendix. Overall, there is a fairly similar distribution of each group

within the site catchment areas although the abundance varies.

Sites that contain evidence of repeated occupation (figure 48) indicate a strong

presence of edible vegetation followed by supplemental and medicinal along the West

Fork and the Elm Fork. However, the Elm Fork values reveal a relatively higher edibility

factor. With respect to change through time (figure 49) similar patterns occur. During

the Middle Archaic and Late Prehistoric I medicinal values rise above supplemental

values along the Elm Fork. The East Fork situation is similar except during the Late

Prehistoric I era medicinal values notably jump above the supplemental values reaching

almost the same level as the edibility values and then gradually decrease. It is likely

that these locations provided seasonal resources that were complimented by a variety

of edible and medicinal plants paired with woodland resources as reflected by

supplemental values. The supplemental value category primarily reflects wood

utilization. As such, sites situated in the Cross Timbers are significantly weighted with

respect to the supplemental category. The same situation is true for sites located along

riparian corridors where woody vegetation tends to aggregate. Fluctuations through

time indicate some type of pattern in relation to the correlation between particular

economic values and site locations. This could be attributed to climate change,

territoriality, seasonality or simply cultural preferences.

193

West Fork Economic Values and Occupation Frequency

0.0350

0.0400

0.0450

0.0500

0.0550

0.0600

SINGLE DUAL TRI MULTI

Edibility Medicinal Supplemental

Elm Fork Economic Values and Occupation Frequency

0.0350

0.0400

0.0450

0.0500

0.0550

0.0600

SINGLE DUAL TRI MULTI

Edibility Medicinal Supplemental

Figure 48. Economic values and occupation frequency.

194

West Fork Economic Values and Change Through Time

0.0350

0.0400

0.0450

0.0500

0.0550

0.0600

ARCH

AIC

EARL

Y AR

CHAI

C

MIDDLE

ARC

HAIC

LATE

ARC

HAIC

LATE

PREH

ISTO

RIC

LATE

PREH

ISTO

RIC

I

LATE

PREH

ISTO

RIC

II

PREH

ISTO

RIC

Edibility Medicinal Supplemental

Elm Fork Economic Values and Change Through Time

0.0350

0.0400

0.0450

0.0500

0.0550

0.0600

ARCH

AIC

EARL

Y AR

CHAI

C

MIDDLE

ARC

HAIC

LATE

ARC

HAIC

LATE

PREH

ISTO

RIC

LATE

PREH

ISTO

RIC

I

LATE

PREH

ISTO

RIC

II

PREH

ISTO

RIC

Edibility Medicinal Supplemental

Figure 49. Economic values and time periods.

195

Mapped distributions of potentials for vegetation types given certain soil combinations

indicate relationships between site locations and concentrations of grains and seed

crops, grasses and legumes, and wild herbaceous plants (figure 50). This figure

illustrates the distribution of soils that have been recoded to represent wild herbaceous

plants, grains and seed crops and medicinal grasses/legumes. The darker values

indicate a lower potential for these types of vegetation whereas the lighter values

indicate an abundance of these resources. Lower values are weighting the sample due

to the lack of soils data where reservoirs now exist. Change through time in relation to

these variables is illustrated in the appendix.

All of the sites in the study area appear to have a stronger relationship with wild

herbaceous plants than any other vegetation group. These types of plants are primarily

associated with forested settings. This relationship is not surprising given that

prehistoric populations appear to have aggregated in the Cross Timbers and in riparian

environments. Grains and seed crops are secondary in relation to change through time

yet tertiary in relation to occupation frequency. This is an interesting relationship that

may indicate that diagnostic sites are found more often in association with this type of

vegetation whereas single occupancy sites (mostly non-diagnostic) are situated in

grassier areas. West Fork sites have a strong correlation with Grasses and Legumes

given that there are sites situated on the ecotone near prairie settings. This type of

vegetation is in the proximity of the Elm Fork sites although it is not concentrated in the

site catchment areas. Overall, relationships between the vegetation types remain fairly

similar with respect to occupation frequency (figure 51) and time period (figure 52).

196

Figure 50. Vegetation potential and prehistoric components.

197

West Fork Vegetation Potential and Occupation Frequency

0.035

0.085

0.135

0.185

0.235

0.285

0.335

0.385

0.435

0.485

SINGLE DUAL TRI MULTI

Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants

Elm Fork Vegetation Potential and Occupation Frequency

0.035

0.085

0.135

0.185

0.235

0.285

0.335

0.385

SINGLE DUAL TRI MULTI

Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants

Figure 51. Vegetation potential values and occupation frequency.

198

West Fork Vegetation Potential and Change Through Time

0.0350.0850.1350.1850.2350.2850.3350.3850.4350.485

ARCH

AIC

EARL

Y AR

CHAI

C

MIDDLE

ARC

HAIC

LATE

ARC

HAIC

LATE

PREH

ISTO

RIC

LATE

PREH

ISTO

RIC

I

LATE

PREH

ISTO

RIC

II

PREH

ISTO

RIC

Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants

Elm Fork Vegetation Potental and Change Through Time

0.0350.0850.1350.1850.2350.2850.3350.3850.435

ARCH

AIC

EARL

Y AR

CHAI

C

MIDDLE

ARC

HAIC

LATE

ARC

HAIC

LATE

PREH

ISTO

RIC

LATE

PREH

ISTO

RIC

I

LATE

PREH

ISTO

RIC

II

PREH

ISTO

RIC

Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants

Figure 52. Vegetation potential values and time periods.

199

Soil potential to provide habitat for certain types of wildlife as estimated by soil survey

data provides a view of another facet of potentiality. Mapped values reveal differential

distribution that indicates that certain types of wildlife are more likely to be

concentrated in particular locations (figure 53). This figure illustrates the distribution of

soils that have been recoded to represent relative potential for rangeland, openland and

wetland wildlife habitat. The darker values indicate a lower potential for these types of

vegetation whereas the lighter values indicate an abundance of these resources. Lower

values are weighting the sample due to the lack of soils data where reservoirs now

exist. Change through time in relation to potential for wildlife habitats is illustrated in

the appendix.

Sites that contain evidence of variable occupation frequency follow similar trends

(figure 54). This is probably due to sample size combined with the fact that sites are

often clustered along the Elm Fork and have overlapping site catchment areas which

influence the outcome. Through time there are fairly consistent correlations between

site location and openland, wetland and rangeland wildlife habitat (figure 55). Wetland

habitat remains fairly low due to the low concentrations of soils associated with these

settings. This is not to say that wetland wildlife was not utilized as much as the other

types. In this case soil classification is weighting the sample. Riverine corridors are not

factored into the distribution of wetland wildlife. Based on the archaeological faunal

data it is known that prehistoric populations, in this region, utilized aquatic resources,

regardless of the time period. Given the limited sample it is difficult to know if this is

the case with all the sites but the proximity to water resources favorably supports this.

200

Figure 53. Wildlife potential and prehistoric components.

201

Elm Fork Wildlife Potental and Occupation Frequency

0.001

0.051

0.101

0.151

0.201

0.251

0.301

0.351

0.401

SINGLE DUAL TRI MULTI

Openland Wildlife Rangeland Wildlife Wetland Wildlife

West Fork Wildlife Potential and Occupation Frequency

0.001

0.051

0.101

0.151

0.201

0.251

0.301

0.351

0.401

SINGLE DUAL TRI MULTI

Openland Wildlife Rangeland Wildlife Wetland Wildlife

Figure 54. Wildlife potential values and occupation frequency.

202

West Fork Wildlife Potential and Change Through Time

0.0010.0510.1010.1510.2010.2510.3010.3510.401

ARCH

AIC

EARL

Y AR

CHAI

C

MIDDLE

ARC

HAIC

LATE

ARC

HAIC

LATE

PREH

ISTO

RIC

LATE

PREH

ISTO

RIC

I

LATE

PREH

ISTO

RIC

II

PREH

ISTO

RIC

Openland Wildlife Rangeland Wildlife Wetland Wildlife

Elm Fork Wildlife Potental and Change Through Time

0.0010.0510.1010.1510.2010.2510.3010.3510.401

ARCH

AIC

EARL

Y AR

CHAI

C

MIDDLE

ARC

HAIC

LATE

ARC

HAIC

LATE

PREH

ISTO

RIC

LATE

PREH

ISTO

RIC

I

LATE

PREH

ISTO

RIC

II

PREH

ISTO

RIC

Openland Wildlife Rangeland Wildlife Wetland Wildlife

Figure 55. Wildlife potential values and time periods.

203

Site Catchment Diversity Sites catchments are measured for relative degrees of diversity based on the

proportional presence of soil series contained in the 1km area. Results (figure 39)

indicate the largest amount of variability in sites that contain evidence of single

occupation. Given the number of sites in this category it is not surprising and in fact

anticipated. However, there are considerably less multi-component sites and this

group also exhibits increased variability.

Figure

parti

al fo

r VIS

ITA

TIO

N

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

VISITATION

Dual

Multi

Single

Tri

56. ANOVA comparisons of site catchment diversity and occupation frequency.

204

With respect to change through time the most significant range of variation occurs

with the Late Archaic sites. The results illustrated by the graph (figure 40) indicate that

there are expressed changes that occur in relation to site catchment diversity. It is

possible that during the Late Archaic populations diversified and enjoyed a greater

range of motion. Previous studies have indicated in prior research that patterns of site

location indicate a decrease in territorial boundaries during the Late Prehistoric. If

these populations were restricted to certain locations under variable climatic conditions

there may be relatively less diversity. However, given the character of the region a fair

amount of diversity is available within the woodlands, prairies, ecotones and riverine

settings such that a certain degree will be continuously present.

parti

al fo

r AG

E

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

AGE

A EA LA LP LPI

LPII

MA P

Figure 57. ANOVA comparisons: Site catchment diversity and time periods.

205

Climatic Interpretations

Climatic conditions are often referred to as explanatory mechanisms for culture

change. Recoded soils data based on pollen records reflects relative vegetal abundance

given various climatic conditions (figure 55). When these data are mapped correlations

Figure 58a. Distribution of climatically abundant vegetation families.

206

can be drawn in relation to site distribution as it changes through time (figure 55 and

figure 56). The working assumption is that populations would have been drawn to

locations that contain abundant opportunities of valuable resources given the

appropriate climatic conditions. This is a generalized model that could be calibrated

with more detailed data. However, due to time constraints it is generalized for the

purpose of illustration. If these patterns are accurate then it is suggested that sites

that were selected by prehistoric populations would have provided a fairly stable

economic resource base in spite of climatic fluctuations. Even though changes through

time are apparent site locations are consistently situated near abundant resources.

Figure 58b. Distribution of climatically abundant vegetation families.

207

Site Activities and Density

Prehistoric activities are inferred based on the types of tools recovered from given

site locations. When they are viewed in relation to the various characteristics of the site

catchments the dynamic inter-relationships between prehistoric activities and the

environment within which they were performed can be evaluated. Sites that contain

evidence of tool manufacturing activities, food preparation, hunting activities and lithic

scatter are predominantly represented in fluvatile and alluvial deposits. In some cases

sites contain evidence of all four activities and as such would be considered to be base

camps. However, the degree of activity that is occurring in a given area can be cross-

checked with the level of occupation density. Correlations indicate at least twelve site

locations exhibit intensive occupation in upland and terrace settings that are associated

with fluvatile and alluvial deposits. Ephemeral sites are primarily correlated with lithic

scatter which are situated in terrace and upland settings situated on fluvatile, alluvial

and sandstone deposits along riparian corridors in the Cross Timbers. Activities

represented at West Fork sites indicate an increased frequency of hunting during the

Archaic and equal distribution of tool manufacture, hunting and food preparation during

the Late Prehistoric. Given that prehistoric sites are non-diagnostic sites defined by

lithic scatter this activity is pronounced for both the West and Elm Forks during the

general Prehistoric. Along the Elm Fork hunting and food preparation are predominant

during the Archaic with a similar pattern accompanied by a slight increase in food

preparation occurring during the Late Prehistoric. Activities in relation to the variables

addressed in this section are summarized by a series of graphs (Figures 59-65).

208

ACTIVITIES: ELM FORK

1015

202530

3540

4550

TOOL HUNTING FOODPREPARATION

LITHIC SCATTER

SIT

E (C

OU

NTS

)

ARCHAIC LATE PREHISTORIC PREHISTORIC

ACTIVITIES: WEST FORK

10

11

12

13

14

15

16

TOOL HUNTING FOODPREPARATION

LITHIC SCATTER

SIT

E (

CO

UN

TS

)

ARCHAIC LATE PREHISTORIC PREHISTORIC

Figure 59. Prehistoric activities and time periods.

Relationships between the number of sites that contain evidence of each particular

activity is illustrated (figure 59). This data is biased by the number of surface sites that

contain limited evidence combined with the abundant data recovered from excavated

sites. It would appear that there is a decrease of activities from the Archaic to the Late

Prehistoric and very little lithic scatter during the Archaic along the Elm Fork. This is

209

not particularly the case. Lithic scatter, for example is a category that includes sites

that contained non-diagnostic lithic scatter. As a result, this category is composed of

non-diagnostic single component prehistoric sites. The same situation is true when

occupation frequency (figure 60) is considered. Future research with a better sample

would help solve this problem.

ACTIVITIES: ELM FORK

10

15

20

25

30

35

40

45

TOOL HUNTING FOODPREPARATION

LITHIC SCATTER

SIT

E (

CO

UN

TS)

MULTI-COMPONENT SINGLE-COMPONENT

ACTIVITIES: WEST FORK

10

11

12

13

14

15

16

17

TOOL HUNTING FOODPREPARATION

LITHIC SCATTER

SIT

E (

COU

NT

S)

MULTI-COMPONENT SINGLE-COMPONENT

Figure 60. Prehistoric activities and occupation frequency.

210

Most of the activities (figure 61) are consistently associated with water deposited

sediments and sandstone geological settings. Repeated testing indicates a preference

for these locations. Shale settings are followed by limestone. A similar pattern emerges

with respect to occupation density (figure 61). Many of the referenced site catchments

contain both sandstone and water deposited sediments such that overlap should be

taken into account.

GEOLOGICAL SETTING: ACTIVITIES

10

20

30

40

50

60

WATERDEPOSITEDSEDIMENTS

LIMESTONESETTINGS

SHALESETTINGS

CLAY RICHLIMESTONESETTINGS

SANDSTONESETTINGS

SIT

ES (

COU

NTS

)

TOOL HUNTING FOOD LITHICS

GEOLOGICAL SETTING SITE: DENSITY

10203040506070

WATERDEPOSITEDSEDIMENTS

LIMESTONESETTINGS

SHALESETTINGS

CLAY RICHLIMESTONESETTINGS

SANDSTONESETTINGS

SIT

ES

(CO

UN

TS)

EPHEMERAL MODERATE INTENSE

Figure 61. Prehistoric activities, site density and geological settings.

211

Patterns related to soil settings (figure 62) indicate a fairly equal distribution of soil

types within catchments that contain various activities and degrees of intensity. The

change between categories appears to be primarily related to the number of sites in

each category. The presence of the various soil settings indicates a strong correlation

with ecotonal settings.

SOIL SETTING: ACTIVITIES

10

20

30

40

50

60

CLAY RICHENVIRONMENTS

SANDYENVIRONMENTS

LOAMYENVIRONMENTS

OTHER

SIT

ES

(C

OU

NTS

)

TOOL HUNTING FOOD LITHICS

SOIL SETTING: DENSITY

102030405060708090

CLAY RICHENVIRONMENTS

SANDYENVIRONMENTS

LOAMYENVIRONMENTS

OTHER

SITE

S (C

OU

NTS

)

EPHEMERAL MODERATE INTENSE

Figure 62. Prehistoric activities, site density and soil settings.

212

great groups (figure 63) tend to produce similar patterns in relation to activities and site

density as did soil settings. The fairly equal distribution appears to be proportionately

affected by the number of sites within each category. In addition, the presence of all

three great groups supports ecotonal utilization regardless of site density or activity.

GREAT GROUPS: ACTIVITIES

10

20

30

40

50

60

MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS

SIT

E (C

OU

NT

S)

TOOL HUNTING FOOD LITHICS

GREAT GROUPS: SITE DENSITY

102030405060708090

MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS

SITE

(C

OU

NTS

)

EPHEMERAL MODERATE INTENSE

Figure 63. Prehistoric activities, site density and great groups.

213

Topographic settings (figure 64) in relation to activities and site density indicate a

variable landscape surrounding site location. Apparently sites are situated in locations

where the catchments include more then one topographic setting. The pattern

continues to proportionately change reflecting utilization of a variety of settings

regardless of activity or site density.

TOPOGRAPHIC SETTING: ACTIVITIES

10

20

30

40

50

60

BOTTOMLANDS WATER(DRAINAGES)

FLOODPLAINS UPLANDS TERRACES

SIT

ES (

COU

NTS

)

TOOL HUNTING FOOD LITHICS

TOPOGRAPHIC SETTING: SITE DENSITY

1020304050607080

BOTTOMLANDS WATER(DRAINAGES)

FLOODPLAINS UPLANDS TERRACES

SIT

ES (

COU

NTS

)

EPHEMERAL MODERATE INTENSE

Figure 64. Prehistoric activities, site density and topographic settings.

214

Given the previous results the ecotonal physiographic setting (figure 65) should have a

higher site frequency. However this graph is a little misleading due to the fact that

ecotonal settings contain a combination of the other categories. This cannot be

avoided due to the site catchment character combined with site distribution. Future

research should delineate sites that are completely within prairie or woodland settings.

The results would probably remain the same due to the character of the region.

PHYSIOGRAPHIC SETTING: ELM FORK

10

1520

2530

3540

45

PRAIRIES CROSSTIMBERS

MAJORRIPARIAN

CORRIDORS

MINORRIPARIAN

CORRIDORS

ECOTONE

SIT

E (

CO

UN

TS)

TOOL HUNTING FOOD LITHICS

PHYSIOGRAPHIC SETTING: WEST FORK

10

20

30

40

50

60

70

PRAIRIES CROSSTIMBERS

MAJORRIPARIAN

CORRIDORS

MINORRIPARIAN

CORRIDORS

ECOTONE

SIT

E (

CO

UN

TS)

EPHEMERAL MODERATE INTENSE

Figure 65. Prehistoric activities, site density and physiographic settings.

215

Statistical Conclusions Results yielded from the statistical analysis indicate that differences between site

locations through time are predominantly due to chance. Similarly there is no indication

of differential selection in relation to occupation frequency. Apparently prehistoric

populations in north-central Texas returned to similar site settings that may vary in

geographical location but not in economic related content. It is important to note that

the statistical results are extremely biased due to the number of sites situated Elm Fork

and the lack of sites along the West Fork of the Trinity. The bias is such that a robust

statistical comparison is not possible until more sites are uncovered. Furthermore,

repeated occupation at sites through time along the Elm Fork heavily weighted

temporal considerations and occupation frequency. The raw environmental and

economic data indicates differential selection. Maps depict this variability in relation to

site catchments. The particular variation between site locations is primarily due to the

different soil settings so there is a considerable difference between sites situated on the

Elm Fork, East Fork and West Fork of the Trinity.

The statistical results (table 8-table 10) are summarized to illustrate the variables

that were depicted and the corresponding results. Although statistical results indicate

that there is not a significant difference between site catchments in relation to

occupation frequency, density, activities or change through time there are a few cases

that do indicate statistically significant differences. Activities and site density indicate

statistically significant differences in relation to change through time and occupation

frequency. This simply means that site function and density varies through time but

216

Table 8. Summary of statistical conclusions.

Variable 1 Variable 2

X-Square df

p-value Statistical

Significance Great groups Components 22.1639 33 0.9239 N Great groups Time 26.2866 84 1 N Great groups C-Time 9.2734 48 1 N Great groups Site Density 13.0649 24 0.965 N Great groups Site Function 13.4503 36 0.9998 N Topography Components 40.1809 45 0.6759 Y Topography Time 49.3356 105 1 N Topography C-Time 19.1622 60 1 N Topography Site Density 19.7875 30 0.922 N Topography Site Function 15.5873 45 1 N Soil Setting Components 60.4014 60 0.4612 Y Soil Setting Time 62.8236 140 1 N Soil Setting C-Time 20.3485 80 1 N Soil Setting Site Density 26.7694 40 0.9459 N Soil Setting Site Function 17.6018 60 1 N Geology Components 34.2261 45 0.8789 N Geology Time 35.4769 105 1 N Geology C-Time 15.3905 60 1 N Geology Site Density 15.3535 30 0.9876 N Geology Site Function 12.4191 45 1 N Physiography Components 21.4053 36 0.9743 N Physiography Time 21.7192 84 1 N Physiography C-Time 8.007 48 1 N Physiography Site Density 9.2363 24 0.997 N Physiography Site Function 8.2276 36 1 N Site Density Components 58.4496 6 0 Y Site Density Time 64.6353 14 0 Y Site Density C-Time 2.4361 8 0.9646 N Site Function Components 32.2334 9 0.0002 Y Site Function Time 41.2895 21 0.0052 Y Site Function C-Time 1.9654 12 0.9995 N Elm Fork and West Fork Combined

not necessarily in relation to site location. Occupation frequency exhibits statistically

significant differences in relation to topography and soil setting. This is the only

indication of differences between site catchments and given the character of the

variables this suggests that there is a pattern operating that can be correlated with site

217

Table 9. Summary of statistical conclusions.

Variable 1 Variable 2 X-Square df p-value ss great groups Sing/Multi (West) 0.5632 3 0.9048 N great groups Sing/Multi (Elm) 1.2972 3 0.9775 N great groups Sing/Multi (Total) 3.3307 11 0.9856 N Topography Sing/Multi (West) 0.6025 4 0.9628 N Topography Sing/Multi (Elm) 0.9917 4 0.911 N Topography Sing/Multi (Total) 5.8985 14 0.969 N Soil Setting Sing/Multi (West) 3.9752 3 0.2642 Y Soil Setting Sing/Multi (Elm) 2.547 3 0.4669 Y Soil Setting Sing/Multi (Total) 5.7574 20 0.9992 N Geology Sing/Multi (West) 0.3855 4 0.9836 N Geology Sing/Multi (Elm) 1.0043 4 0.9091 N Geology Sing/Multi (Total) 2.985 15 0.9996 N Physiography Sing/Multi (West) 0.4115 4 0.9815 N Physiography Sing/Multi (Elm) 1.0354 4 0.9044 N Physiography Sing/Multi (Total) 2.4332 12 0.9984 N Site Density Sing/Multi (West) 4.9184 2 0.0855 Y Site Density Sing/Multi (Elm) 17.1127 2 0.0002 Y Site Density Sing/Multi (Total) 46.1361 2 0 Y Site Function Sing/Multi (West) 2.31 3 0.5106 Y Site Function Sing/Multi (Elm) 9.7762 3 0.0206 Y Site Function Sing/Multi (Total) 38.7188 3 0 Y West and Elm Fork Individually Analyzed

reoccupation. Of all of the variables these two are probably the most informative in

relation to extending interpretation beyond the physical landscape. Along the Elm Fork

of the Trinity there is a statistically significant difference between soil settings with

respect to change through time. Essentially, this means that sites were situated in

areas that would provide access to specific types of vegetation. In addition, there is a

statistically significant difference in site density in relation to soil setting.

218

Table 10. Summary of statistical conclusions.

This i

Setting

there

time in

that

Variable 1 Variable 2 X-Square df p-value ss great groups A/LP/P (West) 1.4139 6 0.965 N great groups A/LP/P (Elm) 1.7195 6 0.9436 N Topography A/LP/P (West) 2.209 8 0.9739 N Topography A/LP/P (Elm) 1.7437 8 0.9879 N Soil Setting A/LP/P (West) 0.257 6 0.9997 N Soil Setting A/LP/P (Elm) 4.0219 6 0.6737 Y Geology A/LP/P (West) 1.3967 8 0.9943 N Geology A/LP/P (Elm) 3.211 8 0.9204 N Physiography A/LP/P (West) 1.4773 8 0.9931 N Physiography A/LP/P (Elm) 1.0117 8 0.9982 N Activities A/LP/P (West) 1.9469 6 0.9245 N Activities A/LP/P (Elm) 49.7815 6 0 Y Activities Soil Setting 1.9554 9 0.9922 N Activities Topography 3.6075 12 0.9895 N Activities Geology 4.5979 12 0.9707 N Activities Physiography 4.55 12 0.9713 N Activities great groups 2.4048 9 0.9833 N Density Soil Setting 3.1109 6 0.7948 Y Density Topography 3.3872 8 0.9078 N Density Geology 4.4548 8 0.8139 N Density Physiography 3.658 8 0.8866 N Density A/LP/P (West) 5.1097 4 0.2762 Y Density A/LP/P (Elm) 67.3799 4 0 Y Density great groups 3.6117 6 0.7291 Y West and Elm Fork Individually Analyzed

ndicates that site density varies according to the particular soil setting. Soil

s tie directly into the edible, medicinal and supplemental vegetation given that

is an association between the two. The fact that site density and change through

correlation with soils settings produce statistically significant results is a signal

prehistoric populations were drawn to specific locations related to the

219

environmental setting. Future hypotheses could be formulated to determine what types

of plants these cultures were drawn to during their respective time periods.

Topographic location provides valuable information in relation to climatic

fluctuations and micro-environmental variation. The statistical significance in this

respect indicates that patterns exist between site location and topographic setting. As

these settings are evaluated it is possible to form explanations for site location in

relation to the climate. As such, this variation is addressed in the final synthesis.

Overall, it can be generally inferred that prehistoric populations consistently selected

site locations for intensive or ephemeral occupation that vary according to soil setting

and topographic setting. These are important clues to discovering the motivation for

such choices. Due to the sample bias posed by the numerous sites along the Elm Fork

that are in close proximity or indicate reoccupation it is necessary to approach the

development of the regional settlement pattern model from a different angle.

One of the benefits of incorporating GIS into SCA is the ability to spatially analyze

site locations with overlay analysis. The GIS interface provides a means to observe the

intersection of the multiple variables, interpret these patterns into dynamic relationships

and observe change through time. In the final synthesis, prehistoric settlement

patterns are observed comparatively between the East, West and Elm Forks of the

Trinity as they change through time by performing an overlay analysis with a GIS

interface. Results from this analysis combined with tabular and descriptive data are

synthesized in a regional model of prehistoric settlement pattern behavior in north-

central Texas.

220

CHAPTER 7

NORTH CENTRAL TEXAS SETTLEMENT PATTERN MODEL

The model presented here is designed to address the character of prehistoric

adaptation to the environment and landscape of north-central Texas. A variety of

factors influence prehistoric decision-making processes. Each of these factors has been

evaluated throughout the course of this thesis. In order to make sense of the pieces

they must be integrated into a coherent whole and synthesized into a regional

settlement model. The primary goal of this research is to formulate theories that would

explain why these populations selected certain locations during their respective time

periods and why these locations were revisited during later time periods. Examination

of the landscape has helped determine the economic value of areas directly associated

with archaeological sites. Evaluation of archaeological remains at these sites has

delineated functional value of these locations. In this chapter sites are comparatively

evaluated and settlement models are developed for each time period within the West

Fork, Elm Fork and East Fork. Settlement patterns that emerge from these

relationships are then evaluated on a regional scale to develop theories regarding

functional site types and adaptations that characterize prehistoric populations. This

model can be utilized to formulate future hypotheses concerning relationships between

these cultures, how they related to each other, to the animals and to the landscape.

Prehistoric Populations along the West Fork of the Trinity

During the Early Archaic sites, along the West Fork of the Trinity, contain evidence

that indicates moderate occupation involving hunting in upland alluvial settings within

221

the Western Cross Timbers. The most intensively occupied site in this portion of the

study area (41WS38) is centrally located in relation to the other sites in the West Fork

drainage. Intensive occupation at this site is indicated by high artifact density, hearth

features, fire cracked rock and faunal remains that have been excavated from midden

deposits. A site of this nature is considered to be a semi-sedentary base camp that was

probably seasonally occupied. The proximity of this multi-component site to the rolling

plains suggests that it was a good location for bands to aggregate and prepare for

communal hunting on the plains. Of all the currently known sites in Wise County this

site appears to be the most frequently occupied based on the diversity of time-

diagnostic projectile points that have been recovered. This site is situated in an

economically advantageous position on the floodplain near a minor tributary. It is

located in a diverse location within the parameters of the Grand Prairie/Western Cross

Timbers ecotone. Economic values associated with this ecotonal locale indicate an

equal presence of edible and medicinal vegetation.

All of the sites that were occupied during the Early Archaic in the West Fork

drainage are also associated with the Middle Archaic time period. An additional site

location is present approximately 30 km to the north in a similar setting. It is located in

an ecotonal setting between the Western Cross Timbers and the Grand Prairie on top of

the same geological foundation but it is on a terrace near a major tributary (Denton

Creek) rather than on a floodplain of a minor tributary. This site is not as intensively

occupied but it has an equivalent economic value. Limited food preparation and

hunting activities are indicated by the artifacts recovered from this site. However, there

222

is no evidence of tool manufacture. Economic potentials are equally distributed

between edible, medicinal and supplemental values. Combined with the frequency of

peripheral non-diagnostic prehistoric sites it is suggested that the Howard site was a

base camp and 41WS7 was an ephemeral hunting camp.

Late Archaic patterns reveal continued site reoccupation of the Middle Archaic sites

with one new site established approximately 30 km to the west and a couple more

about 10km south of 41WS38. Site 41WS43 is situated in a terrace setting on top of

Antlers sandstone near the ecotone that connects the Western Cross Timbers to the

Rolling Plains. The site is ephemeral containing evidence of hunting activities

suggested by projectile points. However, the economic resources offered by the

riparian and ecotonal settings are fairly substantial such that it is likely that it served as

a hunting camp. The overall pattern during this time period suggests a certain level of

sedentism combined with the development of exploratory hunting camps in relation to

the southern sites. Late Archaic sites in the West Fork Study area are less numerous

with relatively lower wildlife and economic values than the previous time periods. The

two base camps are situated near the Grand Prairie/Western Cross Timbers ecotone

whereas the ephemeral hunting camps are in the interior of the Cross Timbers

approximately 5km from the Rolling Plains. This configuration suggests a western

movement from the base camps toward the rolling plains where large game could be

hunted. A westward shift of site locations is the most obvious between the Middle and

Late Archaic throughout the region. It is less noticeable in the West Fork region due to

the paucity of Middle and Late Archaic sites but when viewed in relation to the West

223

Fork this shift is more apparent. Presence of diagnostic points and darts at these

western sites combined with repeated occupation suggests either a migratory route or a

satellite hunting camp. Supporting evidence for a migratory route is provided by an

ephemeral prehistoric site containing non-diagnostic lithic scatter approximately 10 km

west of the nearest diagnostic site and well within the rolling plains. In addition,

although archaeological sites discovered at Lake Bridgeport, 3 km west of this site, have

not been reported to TARL, a multitude of evidence, housed at the University of North

Texas, awaits diagnosis. Sites in Wise County that are in the process of analysis could

add a greater breadth and meaning to this settlement model. Based on the current

settlement model it is expected that sites are located directly west of the study area

and that archaeological evidence reflects cultural affinities to other sites in the region.

Late Prehistoric sites in Wise County are few in number and exhibit opposite

patterns of the previous Late Archaic time period. One could hypothesize based on the

settlement patterns that emerged during this study that population density decreased

along with territorial parameters. During the first phase of the late prehistoric there is

only one site in the West Fork study area that contains evidence associated directly with

this time period designation. Only three sites are associated with the second phase of

the Late Prehistoric and two with the general Late Prehistoric. All of these sites are

occurrences of reoccupation of sites related to previous time periods. As such, it is

difficult to draw any certain conclusions about late Prehistoric populations in the West

Fork study area. However, it would suffice to say that these populations appear to be

retreating from the westward movement indicated during the Late Archaic.

224

Twelve prehistoric sites in Wise County are not associated with any given time

period due to the lack of time-diagnostic artifacts at these locations. However, the

presence of lithic scatter, tools and a variety of other materials that indicate prehistoric

human activity accumulate more meaning when combined with the character of the

landscape. All of these sites with the exception of the southernmost location are

associated with significantly high potential for openland and rangeland wildlife. In

addition edibility medicinal and supplemental values are significantly high at most of

these locations. They are also widely distributed across the landscape unlike the

clustered patterns that emerge in both the Elm Fork and East Fork areas. Given that

the relative distance from 41WS38 ranges between 10km and 30km a return to base

camp within a one day time frame is not likely. If they are related it is likely they were

satellite camps related to specialized activity or migratory camps that facilitated

movement toward the rolling plains to hunt bison and other large game.

Prehistoric populations were widely dispersed in the West Fork portion of the study

area (figure 43). These camps appear to be strategically situated in order to optimize

procurement efforts. Future hypotheses based on this model and corresponding

methodology could predict locations based on the character of the landscape.

Discovery and analysis of additional sites is necessary to further enhance knowledge

concerning prehistoric settlement patterns in the region.

225

Figure 66. Prehistoric populations along the West Fork of the Trinity.

226

Prehistoric Populations along the Elm Fork of the Trinity

Archaeological research has been performed extensively along the Elm Fork of the

Trinity (figure 44). As such, this research is heavily weighted by the increased number

of sites in this portion of the study area. However, it is worth noting that this area

provided a desirable habitat for prehistoric adaptations given the bottomland resources,

mast crop, access to permanent major water resources, ecotonal settings and the

overall diversity of the landscape.

Along the Elm Fork of the Trinity 11 of the 19 Early Archaic sites are aggregated in

a fairly central location along Little Elm Creek. Situated along the northern portion of

the Elm Fork near the Little Elm Creek branch these sites are either on the Woodbine

sandstone formation or on fluvatile terrace deposits. All of the sites in this area are on

the interior of the ecotone between the Western Cross Timbers and the Blackland

Prairie. A high overall economic value indicates a wide range of opportunities including

immediate access to mast crops. Intensity of occupation suggests that Early Archaic

populations recognized this and aggregated in this location. Settlement patterns during

the Early Archaic reveal aggregation in “base camp” locations with hunting, gathering

and/or migratory camps located within 10km to the west and lithic manufacturing

camps 20 km to the south. A wide range of mobility is indicated during this time

period. Gathering activities are complimented in the immediate vicinity of the sites and

the topographic setting is particularly advantageous during the wetter climatic episodes

that occurred during the Early Archaic. All of these factors indicate strategic site

selection.

227

A small aggregate of Early Archaic sites are equally spaced with approximately 1 km

separating them from one another. Collectively these sites form a westward facing

arch that appears to be protecting the confluence of the Elm Fork and Little Elm Creek

from Eastern invaders. These sites also follow the curvature of Running Branch Creek

providing immediate access to riparian resources. The strategic geographical location

of these sites is echoed by their functional presence that alternates between ephemeral

“outposts” and large base camps. These outposts could either signify an off-site

hunting and/or gathering event or sites protecting from northern or eastern

encroachments. On the other hand it is possible that rather than protecting the

confluence to the west these populations were simply following the river system

northward with scouts inspecting the territory upstream. North of these sites near the

confluence of Little Elm Creek a group of sites is situated in a heavily forested

environment in riparian environments. Of all of the sites in the region this group has

the highest frequency of reoccupation and density of artifacts.

There is an interesting inverse relationship between economic value and projectile

point frequency. Two sites in particular in the periphery of the Elm Fork illustrate this

relationship. For example, 41DN62 contains the largest number of artifacts yet a very

low economic value. Conversely, 41DN48 has the lowest number of artifacts and the

highest overall economic value. The wildlife factor is weighting this value although it

also the highest supplemental value and a fairly high edibility value. Hunting activities

are suggested by a combination of factors. The paucity of remains and the fact that it

only contains evidence of single occupation combined with the fact that all of the

228

elements necessary for the hunt including the appropriate habitat for rangeland wildlife

are present support a hunting situation. In addition a more intensively occupied site

(41DN31) is 1.5 km to the southwest that contains lithic debris, midden soil, fire

cracked rock and various other artifacts that suggests a relatively large scale gathering

covering approximately 10 acres. These factors collectively support an Early Archaic

gathering associated with a hunting event and camp near the confluence of Clear Creek

and the Elm Fork of the Trinity. Sherds and celts provide evidence of one other

occupation at site 41DN31 during the Late Prehistoric. Given the size and location of

the site it is very likely that Late Prehistoric people scavenged the site. There are three

reasons why this is thought to be the case. First, the contemporaneous occupation of

the two sites during the Early Archaic combined with the character and proximity of the

sites necessary for fulfillment of the conditions that typically facilitated a communal

hunt. Second, the fact that besides 41DN31 the closest Archaic sites are 7 km away

suggests that this scenario is likely. Third, the paucity of Late Prehistoric artifacts

combined with the size of the site and density of site content suggests that 41DN31

was scavenged for artifacts. The closest Late Prehistoric site is 1.4 km away on the

west side of the Trinity. This site (41DN2) situated on a terrace contains Perdiz,

Bonham and Fresno points, as well as, 20 ceramic sherds. It would be interesting to

compare the sherds from the two sites to evaluate possible similarities. Given the

tendency for territorial tribal conflict, especially during the Late Prehistoric, it is likely

that discovering a previously occupied site stimulated an increased level of

defensiveness. Therefore the selection of the site on a terrace with a river barrier

229

between the two sites appears to be a logical choice of location. This situation is a very

good example of how sites acquire meaning when viewed with respect to occupation

frequency and chronology in relation to other sites in the region.

A clustering of Early Archaic sites situated along the southern portion of Elm Fork

are correlated with fluvatile terrace deposits on the Blackland Prairie side of the Eastern

Cross Timbers ecotonal boundary. 41DN251 and 41DN252 contain Early Archaic

projectile points. These sites, situated in the Blackland Prairie near the confluence of

Denton Creek and the Elm Fork of the Trinity, are less than a kilometer from each

other. Both sites contain evidence of tool manufacture including cores, preforms,

debitage and diagnostic projectile points. Less than a kilometer to the south there are

three Archaic sites that contain non-diagnostic dart points, cores, debitage, preforms,

retouched pieces and fire cracked rock. These five sites are within a 1 kilometer radius

of each other. The closest economic resource is supplemental bottomland hardwoods

that would have provided fuel for the fire to heat treat lithic materials, provide warmth

and cook with. Situated on fluvatile terrace deposits within 2 km of a major tributary

these sites had access to the riparian food resources and protection from any

contemporaneous cultures traversing the River corridor. This cluster of sites appears to

be associated with preparation for the hunt or territorial warfare. A small number of

individuals present at this site are indicated by the moderate density of artifacts. More

than likely it was a band that was in the process of lithic manufacture while collecting

raw materials that were apparently available in the terrace settings of this portion of the

study area. Notably, they are the only Early Archaic sites along the Elm Fork that are

230

located in the Blackland Prairie proper. Once an occupation has been established at a

site it becomes difficult to discern which occupation had the strongest presence.

However, in situations such as this, the frequency of sites in proximity to one another

suggests that a significant amount of activity involving tool manufacture was conducted

in this area during the Early Archaic. The paucity of Late Archaic and late prehistoric

artifacts suggests that repeated visitation during later time periods was likely associated

with recycling of tools and raw materials. While this is not the case in all situations it

appears to be in the northern and southern satellite sites where economic values

associated with the landscape are relatively low within the site catchment. However,

both site clusters are situated near the Elm Fork of the Trinity and small patches of

edible vegetation. Medicinal and supplemental vegetation are also accessible although

not immediately present at the site. One of the sites contains a single Early Archaic

component with light scatter whereas the other site is dual-component and contains

substantial evidence of tool manufacture. Given that the sites are less then 1 km from

each other and contain similar diagnostic projectiles it is likely that there was cultural

interaction between these site locations or that a scavenging event occurred at a later

time. The site that contains considerably more lithic debitage contains evidence of

occupation during both the Early and Late Archaic.

During the Middle Archaic a cluster of sites are focused along Little Elm Creek in site

locations that were pre-established as base camps during the Early Archaic sites.

Besides this cluster there is only one other site located about 5 km south which is also a

case of reoccupation. Situated near the ecotone between the Eastern Cross Timbers

231

and the Blackland Prairie on fluvatile terrace deposits this site primarily contains

evidence of hunting activities that are expressed by a high density of projectile points.

However, the abundance and variety of projectile points indicates some type of tool

manufacture and the presence of cooking pits indicates food preparation. Bison

remains have also been recovered from the site in the form of scapula hoes. Given that

this site was occupied during the Early, Middle and Late Archaic time period it is difficult

to discern which archaeological remains can be attributed to each particular time

period. The main source of differentiation between the time periods is found in the

diagnostic points that are associated with stratified sites in peripheral locales.

During the Late Archaic the Elm Fork site location is once again re-occupied yet to a

much greater degree then previously. Site numbers in areas of reoccupation increase

substantially during the Late Archaic. Pre-established sites to the south and southwest

are also reoccupied. However, during this time period a number of new sites appear in

south, southwest, west and northwest locations. There also appears to be a trend of

movement to the west. For the most part new sites appear in small aggregates.

Choice locations appear to be fluvatile terrace deposits situated in sandstone settings.

All of the new site developments develop in these locations in or near ecotonal settings.

Newly developed sites indicate a notable shift from the Blackland Prairie/Eastern Cross

Timbers ecotone westward to the Grand Prairie/Eastern Cross Timbers ecotone. The

only exception is the southernmost sites situated approximately 5km south of the

camps that were established during the Early Archaic with new peripheral site locations

indicating occupation during the Late Archaic. Activities at these sites appear to be

232

centered on hunting and tool manufacture. 41DN68 and 41DN36 are the exceptions to

the overall pattern that is suggested by Late Archaic sites. These sites are situated

within the Eastern Cross Timbers in upland settings and on fluvatile terrace deposits.

While the overall economic values associated with these sites are fairly low the edibility

values are relatively high. Moving from east to west they are approximately 7 km from

the nearest site indicating that these sites were probably migratory camps. Given the

substantial increase of sites along the Elm Fork of the Trinity there appears to be a

significant population growth during this time period paired with a westward expansion.

Settlement patterns along the Elm Fork of the Trinity with respect to the diagnostic

sites containing Late Prehistoric remains are difficult to discern based on site

reoccupation. During the Late Prehistoric I time period site distribution returns to the

eastern edge of the Eastern Cross Timbers where an ecotone forms with the Blackland

Prairie. Terrace and upland settings in these locations continue to be utilized. During

the Late Prehistoric II time period identical settings are occupied. More information is

available from the distribution of sites that do not contain diagnostic projectiles but

contain untyped arrows and/or ceramics. These sites reveal a continued relationship

with the Grand Prairie/Eastern Cross Timbers eco-tonal boundary and fluvatile terrace

deposits. The ephemeral character of the sites combined with the variety and

abundance of valuable economic resources suggests a correlation with gathering

activities. During the Late Prehistoric populations appear to be growing yet retreating

to previously occupied sites and aggregating into smaller territorial eco-niches.

233

Figure 67. Prehistoric populations along the Elm Fork of the Trinity.

234

Prehistoric Populations along the East Fork of the Trinity

Extensive archaeological research has been conducted in the East Fork portion of

the study area (figure 45) which contains sites that are particularly unique in character.

Given the current state of digital data available for Collin County it was not possible to

evaluate the economic values. Geological information, great groups and topographic

settings are evaluated with respect to site selection. However, the variation is limited

based on the general uniformity of the landscape. All of the sites along the East Fork

are located in the center of the Blackland Prairie. Clay-rich soil characteristic of this

area involve a lot of shrink-swell activity that makes it very difficult for soils to develop.

As a result, besides the predominant tall grass prairie vegetation the main vegetal

resources of food, medicine and supplemental products are contained in the riparian

environments. Site distribution follows this pattern fairly closely.

Very little is known about the Archaic time period. Reportedly only a few sites

were occupied during the Early Archaic. These sites were locations that were later

selected for the construction of Wylie Focus Pits. Whether of not the Wylie Focus Pit

originated during the Early Archaic is unknown but by the Middle Archaic the sites that

would contain these characteristic ceremonial “pits” all provide evidence of prehistoric

activity. Late Archaic populations do not establish new sites but reoccupy sites

established during the Early Archaic. Perhaps these populations had a wider dispersal

across the landscape but were discovered in these locations due to the extensive

excavations that were conducted due to the presence of the Wylie Focus Pits that were

presumably not established until the Late Prehistoric.

235

While some sites suggest occupation during the Archaic time period the bulk of

archaeological remains such as the unusual “Wylie Focus Pit” depressions associated

with hearth features, lithic manufacture, evidence of violent deaths, burials and possible

ceremonial activity does not emerge until the late Prehistoric. This activity probably

corresponds with the increase of population and territoriality that would have put

prehistoric populations in a defensive stance. Supportive evidence for an increase in

populations during the Late Prehistoric is substantiated by the addition of 17 diagnostic

sites along the East Fork of the Trinity and 19 non-diagnostic sites scattered along the

minor tributaries to the east. In relation to the current settlement model significant

patterns emerge from site distribution in the East Fork portion of the study area. These

patterns relate to the centrally based camps that are equally spaced from each other

(8km-10km) across the landscape and then surrounded by clusters of subsidiary sites

that may or may not contain diagnostic material. Whether these sites were hunting

camps, gathering camps, migratory camps or lithic manufacturing camps remains to be

seen. Further research would shed light on the character of these subsidiary sites that

would either support or refute the current settlement model.

The overall character of the base camp sites, artifacts and number of burials

suggests that these sites were inhabited by completely different cultures than those

who inhabited the Elm Fork or West Fork. In addition, the causes of death indicate

conflict and likely conflict of territorial origin. What is difficult to discern is the direction

that the conflict is coming from. Given the character of the sites it seems that Eastern

populations were moving westward where they met with extreme resistance.

236

Figure 68. Prehistoric populations along the East Fork of the Trinity.

237

Functional Site Types in North-Central Texas

Regardless of time period there are common characteristics and functional

differences between sites that prompt the development site types. Five types of sites

are formulated as a result of this research and serve as the core for the current model

of prehistoric settlement patterns. These site types are referred to as base camps,

hunting camps, gathering camps, lithic manufacture camps and migratory camps.

Base camps should contain evidence of extended or multi-component occupation

and be situated in strategic locations that provide increased environmental diversity

with access to optimal economic resources to maximize the gathering contribution of

women and children. These sites should contain higher artifact densities and lithic

manufacture as it is not economic to transport large amounts of lithic materials during

the hunt. Tool manufacturing debris, food preparation artifacts, and food storage

containers are associated with base camps. Abundance of projectiles indicates work

stations at base camps.

Hunting camps are sites that contain evidence of ephemeral single visitation are

situated in strategic locations where hunting would be optimized. In particular, these

sites are associated with areas that have a higher potential for wildlife resources. As

such, these sites provide access to economic resources that may be more specialized in

terms of value relative to sites that contain evidence of higher density occupation. Sites

with low artifact density may have been utilized as food-processing stations where the

hunted animal was prepared for transport and dull or excess flakes discarded on site.

At these site locations there may also be evidence of tool maintenance, a logical by-

238

product required for the process of preparation. Moderate artifact diversities indicate

hunting camps depending on the character and abundance of the remains.

Gathering camps are sites that contain evidence of ephemeral or repeated

occupation combined with a low artifact density. This site type acquires meaning in

relation to the relative economic value. It is expected that at these particular sites

there is a high economic value for edible, medicinal, supplemental resources or a

combination therein. In order to be categorized as a camp the site location must have

access to a water source.

Lithic manufacture camps are sites that contain artifacts that are primarily and

directly associated with lithic manufacture. In addition, these sites if there are to be

seen as specialized activity sites will be located near areas that would provide

immediate access to lithic raw materials. It is also expected that such sites will contain

heat-treated or fire cracked rocks and in a location that has access to wood bearing

resources. While edible vegetation would be of secondary importance water must be

accessible within a reasonable distance from the site.

Migratory camps are ephemeral sites that contain similar archaeological evidence to

food procurement camps. This site type takes on meaning in association with other

sites that are similar in character and indicate a directional movement toward a territory

that would provide an economic resource that is seasonally available. In addition, these

sites will demonstrate a geographical or archaeological relationship with specific base

camps. While these sites are not expected to be associated with high economic values

they are expected to be situated in a location that has access to a water resource.

239

Prehistoric Adaptations

Prehistoric adaptations in north-central Texas vary according to the relative

geographical and ecological setting. Similarities and differences between the relative

settlement patterns have emerged during the course of this research. Factors such as

geology, soils, physiographic setting, topographic setting, climate, economic resource

base and population density influence the range of necessary adaptations.

Early and Middle Archaic cultures in north-central Texas set the stage and

established the foundations for future adaptations in this region. This is mainly due to

the natural circumstances they had to adapt to, as the landscape responded to the

changing conditions, during the transition from late glacial to post glacial climates.

Essentially, humans had to adjust to the establishment of extensive forests that

accompanied this transition. As such, it is expected that sites would be situated near

bottomland resources. This appears to be the case, particularly during the Early Archaic,

based on a combination of the faunal remains and geographic distribution of sites.

Faunal remains and the geographic distribution of sites also indicate that these low

density populations enjoyed a fairly wide range of mobility. However, sites associated

with the Early and Middle Archaic tend to aggregate in characteristic locations, with a

few exceptions, which can be explained by specific cultural adaptations. Lynott (1977)

observes a shift from open-land big game hunting to bottomland riverine hunting and

gathering during the Early and Middle Archaic. According to faunal evidence,

topographic and physiographic settings this appears to be the case. McCormick et al.

(1975) correlate Archaic seasonal camps with ecotonal settings and secondary activity-

240

specific sites spread out across the landscape strategically. It has been observed

during the course of this research that sites are primarily situated in ecotonal settings

with few exceptions. These exceptions find logical explanations by correlating the site

contents with the character of the landscape. For example, the correlation of lithic

manufacturing stations in association with geological settings, that would provide raw

materials, is a common situation.

On a regional scale something happened between the Middle and Late Archaic that

prompted a large scale westward expansion. The reason why a migration is

hypothesized is due to the character of the newly established. Sites that were

considered to be a part of this westward expansion were single occupation ephemeral

sites situated between 10-20 km from the “base camp” sites. The direction toward the

Plains is a probable migration route given that populations are known to have migrated

to the Plains during certain time periods to hunt bison. It is also known that trade with

the Plains Indians and New Mexico Pueblos occurred. Although, it is possible that

populations were expanding their site distribution away from the more desirable riverine

basins due to population increase, there is no evidence that indicates that they were

moving their base camps. Given the dry climatic conditions of the previous time period

and the wetter climate experienced by Late Archaic cultures it is likely that the

population increased and that they were expanding their hunting territory. There is

clearly an increase in the number of sites between the time periods which does suggest

an increase in population. Lynott (1977) observes an increase in population and

decrease in territoriality. Site distribution and frequency combined with the climate

241

patterns supports this. The wetter conditions provided more abundant resources so it is

not likely that they migrated out of the region. It is possible that they were returning to

the region after the Middle Archaic drought. The paucity of sites in the region during

the Middle Archaic suggests that populations were responding to drought conditions by

migrating to other locales so it is highly likely that they would have returned to their

previously established base camps when conditions permitted. The character of the

newly established sites is ephemeral and suggests hunting activities by the presence of

diagnostic point types. The complexity introduced by these sites is their distance from

any peripheral base camps and the appearance of all of these sites to the west of the

majority of established base camps. This is a question that can only be answered by

future research and the discovery of additional base camps.

Typically the initial pattern established by Archaic cultures involves a centrally

located semi-sedentary base camp surrounded by ephemeral camps that facilitated

hunting, gathering, migration or lithic manufacture. Besides the functional site types,

that appear to be re-established through time, regional behavior is discernible by the

geographical distribution of sites. Archaic cultures have a much wider range of

occupation then late prehistoric cultures. This is a phenomenon that requires more

research with respect to environmental and cultural resources. There is more than one

reason. Perhaps it is more straightforward for the late prehistoric but for the Archaic it

is not just a question of territoriality. It is a question that has to do with why Archaic

cultures chose to be highly mobile when they could have subsisted just as other

cultures did during other time periods under the same environmental conditions.

242

Late prehistoric cultures in north-central Texas walked upon common ground and

the knowledge passed down from their ancestors. These populations were drawn back

to the same locations that were utilized by the Archaic cultures. Base camps were re-

established and hunting/gathering camps were either revisited or re-established in

similar environments where economic values facilitated such activity. Lithic

manufacturing stations probably were established in locations where raw materials were

available naturally or where previous occupants left workable lithic debris behind. Tools

were likely recycled, replicated or refashioned. Cumulative knowledge concerning the

proper use of medicinal, edible and supplemental vegetation probably opened doors to

greater variety. However, paired with the knowledge passed on from previous

generations, population growth and pressure created new problems. Climatic

conditions would have influenced the success of the harvest of a seasonal plant or the

migratory patterns of animals that provided sustenance for these populations. During

the second phase of the Late Prehistoric the effects of drought conditions reappear

along with the utilization of bottomland resources, decrease in aquatic resources and

increase in the variety of fauna utilized during this time period. Soil character and

degree of development certainly provided variable opportunities as late prehistoric

populations turned to horticulture. In some cases the physiographic setting provided

more opportunities and in others restrictions or barriers. Topographic settings appear

to respond to climatic conditions and more pronounced territorial issues.

Regardless of territorialism, late prehistoric cultures still found ways to situate

themselves in their “preferred” locations for extended amounts of time. If they were

243

out of their “comfort zone” it does not appear that they were for very long. Even

though the population certainly increased, late prehistoric populations tend to occupy

sites that were previously occupied by the Archaic, with a tremendous decrease in

territorial range. However, it should be noted that there are many unknowns with the

number of “non-diagnostic” prehistoric sites present in the region. This pattern is

probably more related to the character of the research conducted in the region than it

is a real reflection of the behavior of late prehistoric cultures.

Regardless of the time period or climatic conditions prehistoric preferences in

relation to site selection appear to be fairly consistent through time. As Bousmann and

Verrett (1973) suggested fluviatile terraces often function to accommodate base camps.

Also, as McCormick et al. (1975) pointed out, specialized activity locations are often in

locations that are amenable to the activity. Several investigators emphasize the

importance of ecotonal settings. Ecotones are predominantly selected to provide access

to food and shelter. The importance of ecotonal settings is supported by the

archaeological faunal evidence and the placement of the sites on the landscapes.

Riparian environments and sites situated with reasonable access to water is often

combined with the benefits provided by the ecotonal setting.

During the course of this research interesting patterns and fascinating questions

have emerged. Interestingly, it is evident that through time based camps were re-

established and ephemeral camps distributed either in the near proximity of these base

camps or in a wider scattered dispersal around the heart of the base camp sites. In

these cases reoccupation of ephemeral sites appears to be based more on random

244

chance then on some type of learned pattern. Reoccupation can be better explained

with the addition of an interpretable landscape expanding the range of possibilities that

in combination with the material remains could help explain site selection. This

ecological approach needs to be expanded and refined with additional environmental

data collection and evaluation. Future research designed to address regional questions

such as reoccupation and variable dispersal is the only way that these patterns will

come to light.

A pattern of rhythmic dispersal and return appears to be the central organizational

mechanism underneath prehistoric human behavioral choices in north-central Texas.

Only future research can determine why this pattern exists and if indeed history repeats

itself through a natural process, some sort of implicit instinct, that revolves around the

need to sustain human life. As a result, one must ask oneself if what happened in the

past becomes a part of our inherited inherent knowledge that is utilized to survive in

the present. This is a question that I think archaeological research was initially

designed to answer but unless all of the parts of the whole are organized, preserved,

shared and understood this question will remain unanswered. With the knowledge

provided by the current regional settlement model and the accompanying methodology

it is possible to conduct future research to test a number of hypotheses that question

site function, distribution and large scale regional events such as the Middle to Late

Archaic westward expansion. These more specific hypotheses can provide clues that

when viewed collectively will eventually lead to answers for the larger more abstract

questions that science continually seeks to answer.

245

CHAPTER 8

DISCUSSION AND FUTURE RECOMMENDATIONS

Throughout the course of this research ecological, geographical and archaeological

methods have been evaluated, cultural histories reviewed and a new innovative

research methodology formulated to facilitate future research and enhance

interpretations of prehistoric behavior in north-central Texas. In order to build a

regional settlement model SCA has been conducted to extract and interpret

environmental information to shed light on the character of archaeological sites in

combination with previously excavated remains. Geographic Information System (GIS)

software has served as a useful tool for illustrating, comparing, organizing, storing and

explaining observations on a regional scale.

Archaeologists concerned with the dynamics of prehistoric behavior must take into

consideration the dynamics of the ecological system. Given the evidence available for

north-central Texas the best one can hope for on a regional scale is to develop an

understanding of the dynamics of the micro-environments around site locations,

correlate this information to the material remains, relate it to the larger network of sites

and draw inferences by comparison. Framing the archaeological remains in an

ecological context, addressing a series of research questions based on ecological and

archaeological data and analyzing spatial patterning to formulate a regional model is

the basis of the proposed methodology. The methodology can be refined, reshaped

and extended to fit the particular character of the region where it is employed. What is

of primary importance is that basic conditions are met to fulfill multiple research efforts.

246

Essentially, what this means is that site catchment analysis (SCA) and all that this

phrase implies should be incorporated into archaeological methodology.

Essentially, learning is the basis of a particular manifestation of cultural behavior. In

order to understand the mechanism of cultural selection, the process by which new

behaviors are taught, learned and retained should be investigated. For example, it

must be taken into consideration that food procurement techniques were much more

dependent upon geographical location during prehistoric times than they are today and

as such learning would have been as intricately intertwined with spatial location. The

dynamic character of this cultural process makes it very difficult to discern in the

archaeological record but the environment provides clues that can help understand the

larger picture. In addition, a deeper understanding of the resources available to ancient

populations has shed light on necessary decisions associated with that specific type of

environment. For example, many medicinal plants are highly toxic and potentially lethal

if not prepared in an appropriate manner. Therefore, a medicinal environment requires

a certain degree of knowledge of what plants to choose and how to prepare these

plants. Another instance of learning relates to cultural affinities related to style which

could either be due to trading goods or exchanging knowledge. On the other hand, it is

often the case that flint-knappers discovers a particular tool type and based on their

knowledge of tool manufacture replicate the item without outside instruction. As such,

it is very difficult to ascertain the basis of cultural affinity and to what extent it is a

result of primary or secondary case of information diffusion. In this case examination

of the artifacts in isolation could result in the possibility of either case. However, it is

247

possible to establish theories when other variables are factored into the equation.

Other artifacts, faunal remains, palynological remains and possibly structural remains all

provide clues that shed light on cultural patterns. If a solid correlation is established

between two particular sites it is possible to look for processes and reciprocity.

In addition to the complexity of social systems the complexity of ecosystems poses

a problem in and of itself in terms of obtaining a database that is powerful enough to

reconstruct past systemic interactions between humans and their landscape. Vayda

and Rappaport (1968: 495-496) caution against the application of an ecological

approach by emphasizing its limitations (cited in Geier, 1974: 47). Regardless of the

limitations that accompany any scientific inquiry into the past it is beneficial to begin

constructing databases that include consideration of the components of the ecosystem

and their potentiality to exist in the past. Although the task may seem insurmountable

it is a worthwhile venture that provides a more comprehensive view of the potentiality

of past cultural systems. It is true, as Vayda and Rappaport (1968) point out, that most

statements of socio-environmental interaction are hypotheses only. However, this is

the case with any other inference that archaeologists make about the past. The

archaeological record presents “knowns” and “unknowns.” It is what we do with the

“knowns” that will determine the depth and breadth of our learning experience.

Future Recommendations

Destruction of archaeological sites by agriculture, urban expansion, highway

construction, reservoir development projects and a variety of other federal, state, and

locally funded construction projects is a continual phenomenon. Mitigation of this

248

problem has been attempted by the passing of legislation such as the National

Environmental Policy Act of 1969. The inclusion of cultural resources as a significant

part of environmental impact assessments has provided archaeologists with the

opportunity to become involved in research and planning of land development to a

certain degree. Legislation protecting archaeological resources has produced massive

data collection in the archaeological community. Coupled with the lack of

communication between state and local agencies this situation has led to increased

fragmentation in the archaeological record.

Archaeology as a discipline has only been academically active for a little over 50

years. As a relatively young discipline, experimentation with research methodology and

excavation techniques is still in progress. In general, archaeology has come a long way

since the antiquarian days of collecting and early pre-conceived anthropological notions

of aboriginal cultures. Inter-disciplinary studies have contributed a wealth of

information to help explain prehistoric behaviors. Cultural ecology, ethnography and

ethnobotany, to name a few, have provided a deeper understanding and appreciation

for the necessary intellect required for survival during prehistoric times.

We no longer think of the pre-ceramic plant-collectors as a ragged and scruffy bands of nomads; instead they appear as a practiced and ingenious team of lay botanists who know how to wring the most out of a superficially bleak environment (Flannery, 1968:67).

A plethora of archaeological methods have been innovated since the inception of

archaeology as an academic discipline. These methods have provided focused efforts

and increased efficiency in terms of time management while contributing pertinent and

valuable information to the archaeological record. However, the associated data and

249

information becomes lost in the shuffle along with the exclusion of data from the site

context that could provide important data for other investigations. As a result, many

scientific inquiries become redundant procedures and comparability is limited to the

research methodology under which the initial excavations were conducted.

Archaeologists have been so focused on the particulars that the larger picture is either

obscured or over-generalized. While attempting to establish a regional model much

time has been consumed by re-entry of data and information. In addition, there is a

significant amount of data exclusion due to the unavailability of resource information.

It is very difficult for growth to occur in isolation. Archaeology should be a cumulative

discipline that is organized and accessible. Regardless of the situation in the

archaeological community land development continues and site reports accumulate in

inaccessible locations. What is lacking in this system of piecemeal cultural resource

management efforts is a standardized methodology to guide local archaeological

societies, academic institutions, archaeological consultants and consulting firms. While

these entities are due creative process, there needs to be an established set of

minimum requirements in place for data collection and reporting, in order for

excavation efforts to benefit future research and facilitate comparability. In addition,

there needs to be a mechanism in place to centralize and organize these data for future

investigations. Growth in technology over the past five years has provided the means

to make this happen. GIS technology provides the perfect organizing mechanism that

is well suited to the character of archaeological data. The ultimate goal is the creation

and maintenance of centralized regional databases, delineated by physiographic

250

boundaries, that links the archaeological data to spatial location and interpretive

information. The construction of a database of this nature can facilitate a variety of

archaeological, statistical and spatial analyses. With this in mind, the current research

was designed to create a proto-type of a regional archaeological database that was

incorporated into a GIS interface and linked to an interpretive environmental database.

SCA was the established research methodology selected to create a model of prehistoric

settlement patterns in north-central Texas. The design is such that multiple analyses

could be performed and various research methodologies employed. At this point in

time there is a necessity to organize the data that has been collected so that it can be

accessed and synthesized for cumulative analyses. In order to fulfill these goals a

mechanism must be in place to organize our prior knowledge so that it can be used

analytically. Otherwise we will continue to piece together a past that is biased by the

available data that has been collected to fulfill particular research methodologies and

the record will remain as fragmented in reports as it is in the ground. There is

assuredly data that is irretrievable. However, simply organizing what we already have

and establishing standards for future data collection is a step in the right direction.

Future excavations should include analyses beyond the archaeological remains including

active field surveys and excavation of environmental information.

1. Systematic excavation of the site with particular attention to the recovery of economic data

2. Identification of floral and faunal species, and soil character, recovered from

cultural deposits

3. Delineation and description of the microenvironmental habitat of each species

251

4. Field survey of the environment in the region of the site including topographic features, available resources, and cultural limitations of access within a 1-2 km radius of the site location. This would require extensive field survey, test units to collect natural pollen and micro-faunal samples with stratigraphic time- depth, soil samples and vegetation samples of the modern natural environment.

5. Delineation of concentric temporal contours around the site representing travel

times from the site to facilitate interpretation and field strategy 6. Description of technological aspects of the culture represented at the site, with

particular reference to patterns of subsistence and settlement based on the assemblage character.

7. Identification of the number and proportion of the resources found in the deposit

which could have been obtained from each of the several zones as defined by the temporal contours

8. Determining the zone (site territory) which can account for procurement of the

respective resources.

During the Stone Age, Bronze Age, Iron Age, Medieval, Renaissance and the Industrial

age in the Old World, archaeological evidence suggests that hunter-gatherers in north-

central Texas enjoyed the comforts of a landscape structured only by plants, animals,

and other humans. For over twelve thousand years people lived in this region and the

only traces of their existence are in the archaeological record, as artifacts, features, and

concentrations of other types of archaeological materials. Without these records, from

our perspective, in time and space, they did not exist. The challenge, then, is to record

the existence of these prehistoric populations with as much detail as possible so that we

can learn from the behavior of our ancestors to create a better future. We live in an

Information age where data surpasses our current ability to deal with it adequately but

we can begin fashioning and structuring databases for the archaeological record with

integrated systems that will archive this information for future generations.

252

APPENDIX

253

Figure A1. Wildlife potentials and prehistoric sites.

254

Figure A2. Wildlife potentials and Archaic projectile points.

255

Figure A3. Wildlife potentials and Early Archaic projectile points.

256

Figure A4. Wildlife potentials and Middle Archaic projectile points.

257

Figure A5. Wildlife potentials and Late Archaic projectile points.

258

Figure A6. Wildlife potentials and Late Prehistoric projectile points.

259

Figure A7. Wildlife potentials and Late Prehistoric I projectile points.

260

Figure A8. Wildlife potentials and Late Prehistoric II projectile points.

261

Figure A9. Vegetation potentials and prehistoric sites.

262

Figure A10. Vegetation potentials and Archaic projectile points.

263

Figure A11. Vegetation potentials and Early Archaic projectile points.

264

Figure A12. Vegetation potentials and Middle Archaic projectile points.

265

Figure A13. Vegetation potentials and Late Archaic projectile points.

266

Figure A14. Vegetation potentials and Late Prehistoric projectile points.

267

Figure A15. Vegetation potentials and Late Prehistoric I projectile points.

268

Figure A16. Vegetation potentials and Late Prehistoric II projectile points.

269

Figure A17. Economic potentials and prehistoric sites.

270

Figure A18. Economic potentials and Archaic projectile points.

271

Figure A19. Economic potential and Early Archaic projectile points.

272

Figure A20. Economic potentials and Middle Archaic projectile points.

273

Figure A21. Economic potentials and Late Archaic projectile points.

274

Figure A22. Economic potentials and Late Prehistoric projectile points.

275

Figure A23. Economic potentials and Late Prehistoric I projectile points.

276

Figure A24. Economic potentials and Late Prehistoric II projectile points.

277

Table A1. Raw data for hypothesis testing.

EAST FORK OCCUPATION FREQUENCY CHANGE THROUGH TIMEGREAT GROUPS SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII CHROMUDERTS 44 2 1 1 19 13 1 2 1 18 3 16 CHROMUSTERTS 27 0 1 0 13 6 0 1 0 10 1 9 HAPLAQUOLLS 37 2 1 1 16 10 1 2 1 17 3 15 HAPLUSTOLLS 32 2 1 2 12 13 1 3 1 15 4 13 OCHRAQUALFS 17 1 0 0 11 3 0 0 0 3 0 3 PALEUSTALFS 1 0 0 0 1 0 1 0 0 0 0 0 PELLUSTERTS 45 2 1 3 17 16 2 4 2 2 5 20 RENDOLLS 1 0 0 0 0 0 0 0 0 1 0 1 USTOCHREPTS 35 2 1 3 14 14 2 4 2 17 5 15 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20 OCCUPATION FREQUENCY CHANGE THROUGH TIMETOPOGRAPHIC SETTING SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII ALLUVIAL TERRACES 45 2 1 3 17 16 2 4 2 22 5 20 FLOODPLAINS 37 2 1 1 16 10 1 2 1 17 3 15 UPLANDS 44 2 1 1 19 13 1 2 1 18 3 16 UPLANDS, STREAM TERRACES 32 2 1 3 10 14 2 4 2 18 5 16 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20 OCCUPATION FREQUENCY CHANGE THROUGH TIMEGEOLOGICAL SETTING SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII ALLUVIUM 36 2 1 1 15 9 1 2 1 18 3 16 AUSTIN GROUP 5 0 0 0 0 2 0 0 0 4 0 4 FLUVATILE TERRACE DEPOSITS 21 1 0 3 9 9 2 3 2 9 3 9 MARLBROOK MARL 4 0 0 0 4 0 0 0 0 0 0 0 OZAN FORMATION 26 2 1 0 9 7 0 1 0 14 2 12 PECAN GAP CHALK 10 0 0 0 10 0 0 0 0 0 0 0 WOLFE CITY SAND 7 0 0 0 7 0 0 0 0 0 0 0 WATER 14 0 0 2 7 5 1 2 1 6 2 6 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20 OCCUPATION FREQUENCY CHANGE THROUGH TIMEPHYSIOGRAPHIC REGION SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII BLACKLAND PRAIRIE 47 2 1 3 19 16 2 3 2 22 5 20 RIPARIAN ZONE (EAST FORK) (2KM) 13 0 0 1 0 6 1 1 1 9 1 9 RIPARIAN ZONE (MINOR TRIB 2KM) 24 2 1 1 8 7 1 2 1 15 3 13 EPHEMERAL STREAMS (1KM) 13 0 1 0 9 2 0 1 0 4 1 3 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20

278

Table A2. Raw data for hypothesis testing.

ELM/WEST FORKS OCCUPATION FREQUENCY CHANGE THROUGH TIMEGREAT GROUP SING DUAL TRI MULTI P A EA MA LA LP LPI LPII ARGIUSTOLLS 5 2 1 1 4 2 1 2 4 2 2 2 CALCIUSTOLLS 28 3 5 5 21 9 6 6 15 11 7 6 HAPLUDERTS 31 10 4 3 29 10 7 3 13 15 4 4 HAPLUSTALFS 57 13 7 12 46 21 17 10 27 29 12 17 HAPLUSTEPTS 39 17 4 10 33 16 14 9 23 28 11 13 HAPLUSTERTS 48 11 8 10 38 18 16 8 24 28 12 15 HAPLUSTOLLS 45 14 6 11 42 17 13 9 26 27 12 15 PALEUSTALFS 70 20 8 12 58 26 19 10 33 35 12 17 PALEUSTOLLS 15 2 1 0 12 3 2 0 2 2 0 1 RHODUSTALFS 3 0 0 0 3 0 0 0 0 0 0 0 USTIFLUVENTS 34 11 4 5 28 16 10 4 16 16 2 8 TOPOGRAPHIC SETTING SING DUAL TRI MULTI P A EA MA LA LP LPI LPII BOTTOMLANDS 26 9 4 4 20 14 9 3 14 15 2 6 FLOODPLAINS 55 17 7 12 47 22 16 10 30 31 12 17 LOW HILLS AND RIDGES 50 11 8 9 36 19 14 6 26 30 11 13 OLD TERRACES AND VALLEY FILLS 2 1 0 1 2 1 1 2 2 1 1 1 PLAINS 6 3 0 2 6 3 2 3 4 2 1 3 PLEISTOCENE AGE TERRACES 27 4 7 8 17 13 11 6 18 21 10 12 RIDGES 11 2 0 3 11 3 3 4 4 4 1 3 STREAM DIVIDES 9 0 0 1 9 1 0 0 1 1 1 1 STREAM TERRACES 57 14 7 11 48 22 15 10 26 27 11 16 TERRACES 41 7 6 8 30 13 12 6 20 22 10 12 UPLAND RIDGES 13 1 0 3 13 3 2 3 4 4 2 2 UPLANDS 68 20 8 12 59 44 32 15 55 57 21 26 TERRACES, FOOTSLOPES, FANS 28 15 4 8 23 14 12 6 20 25 10 10 GEOLOGICAL SETTING SING DUAL TRI MULTI P A EA MA LA LP LPI LPII ALLUVIUM 40 11 7 8 33 19 13 7 23 22 7 11 ANTLERS SAND 9 2 0 0 9 1 0 1 2 0 0 0 BOKCHITO FORMATION 2 0 0 0 2 0 0 0 0 0 0 0 EAGLE FORD FORMATION 10 5 1 2 8 3 4 2 4 7 3 4 FLUVATILE TERRACE DEPOSITS 36 14 5 9 31 17 12 7 22 25 9 13 GLEN ROSE LIMESTONE 1 0 0 0 1 0 0 0 0 0 0 LIMESTONE AND CLAY 6 1 0 1 6 1 1 2 2 1 1 1 MARL AND LIMESTONE 2 0 0 1 2 1 0 0 1 1 1 1 KIAMICHI FORMATION 1 0 0 0 1 0 0 0 0 0 0 0 PALUXY SAND 1 0 0 2 1 2 2 2 2 3 1 1 PAWPAW FORMATION 4 0 0 0 4 0 0 0 0 0 0 0 SURFICIAL DEPOSITS 1 0 0 1 1 1 1 1 1 2 0 0

279

Table A3. Raw data for hypothesis testing.

ELM/WEST FORKS OCCUPATION FREQUENCY

CHANGE THROUGH TIMESOIL SETTING SING DUAL TRI M P A EA MA LA LP LPI LPII BLACKLAND 37 8 6 9 29 13 13 7 20 21 11 13 CLAY LOAM 43 17 5 9 35 17 14 8 23 28 11 13 CLAYEY BOTTOMLAND 36 11 7 6 30 15 11 6 21 21 7 8 CLAYPAN PRAIRIE 52 15 6 11 43 20 15 9 26 27 12 16 DEEP REDLAND 6 0 0 0 5 1 0 0 0 0 0 0 DEEP SAND 14 2 4 6 11 8 8 8 13 10 5 7 ERODED BLACKLAND 15 5 2 5 11 4 7 3 9 15 7 7 LOAMY BOTTOMLAND 56 17 8 12 48 24 17 10 31 32 12 17 LOAMY PRAIRIE 0 1 0 0 0 1 0 0 0 0 0 1 LOAMY SAND 24 6 4 8 17 12 10 7 15 18 6 12 LOW STONY HILL 2 0 0 0 2 0 0 0 0 0 0 0 REDLAND 4 1 1 0 3 0 0 1 3 1 1 0 ROCKY HILL 1 0 0 0 1 0 0 0 0 0 0 0 SANDSTONE 3 1 0 0 3 1 0 0 1 0 0 0 SANDY LOAM 69 20 8 12 57 26 19 10 33 35 12 17 SANDY 19 2 3 4 11 12 5 2 11 8 3 6 SHALLOW CLAY 2 1 0 2 2 2 1 1 2 2 1 3 SHALLOW 1 0 0 0 1 0 0 0 0 0 0 0 STEEP ADOBE 8 1 0 1 7 1 1 2 2 2 1 1 TIGHT SANDY LOAM 11 3 0 3 11 4 3 4 5 4 1 3 WATER 61 16 8 11 51 23 18 9 30 32 11 16

ELM/WEST FORKS OCCUPATION FREQUENCY

CHANGE THROUGH TIMEPHYSIOGRAPHIC REGION SING DUAL TRI M P A EA MA LA LP LPI LPII BLACKLAND PRAIRIE 13 7 1 0 14 4 3 1 4 2 1 1 EAST CROSS TIMBERS 49 11 8 9 36 19 14 6 26 29 11 13 GRAND PRAIRIE 22 4 3 2 17 7 3 2 9 10 2 3 ROLLING PLAINS 1 0 0 0 1 0 0 0 0 0 0 0 WEST CROSS TIMBERS 11 3 0 3 11 4 3 4 5 4 1 3 RIPARIAN ZONE (ELM FORK) (2K) 16 1 3 0 11 4 3 0 4 5 0 1 RIPARIAN ZONE (WEST FORK) (2K) 2 0 0 1 2 1 1 1 1 2 0 0 RIPARIAN ZONE (DENTON CREEK) (2K) 3 1 0 0 4 0 0 0 1 0 0 0 RIPARIAN ZONE (MINOR TRIBUTARIES) (2K) 28 4 4 7 19 12 9 5 16 19 8 8 EPHEMERAL STREAMS (1KM) 20 4 4 4 15 12 6 3 11 11 3 4 ECOTONE (B PRAIRIE/E CROSS TIMBERS) (2K) 5 0 1 0 3 1 1 1 2 1 1 0 ECOTONE (E CROSS TIMBERS/G PRAIRIE) (2K) 14 3 3 1 10 6 2 0 7 8 1 2 ECOTONE (G PRAIRIE/W CROSS TIMBERS) (2K) 6 1 0 1 6 1 1 2 2 1 1 1 TOTAL 71 16 8 11 59 26 19 10 33 35 12 17

280

Table A4. Raw data for hypothesis testing.

ELM/WEST FORKS OCCUPATION DENSITY

GREAT GROUP EPHEMERAL MODERATE INTENSIVE TOTAL ARGIUSTOLLS 5 2 2 9

CALCIUSTOLLS 25 6 10 41

HAPLUDERTS 29 12 7 48

HAPLUSTALFS 55 21 13 89

HAPLUSTEPTS 39 18 13 70

HAPLUSTERTS 45 18 14 77

HAPLUSTOLLS 43 19 14 76

PALEUSTALFS 67 28 15 110

PALEUSTOLLS 14 3 1 18

RHODUSTALFS 3 0 0 3

USTIFLUVENTS 30 18 6 54

TOPOGRAPHIC SETTING EPHEMERAL MODERATE INTENSIVE TOTAL BOTTOMLANDS 22 15 6 43

FLOODPLAINS 52 24 15 91

LOW HILLS AND RIDGES 48 18 12 78

OLD TERRACES AND VALLEY FILLS 2 1 1 4

PLAINS 6 4 1 11

PLEISTOCENE AGE TERRACES 26 9 11 46

RIDGES 10 4 2 16

STREAM DIVIDES 9 0 1 10

STREAM TERRACES 53 23 12 88

TERRACES 39 13 10 62

UPLAND RIDGES 12 2 3 17

UPLANDS 66 28 15 109

TERRACES, FOOTSLOPES, ALLUVIAL FANS 29 15 11 55

GEOLOGICAL SETTING EPHEMERAL MODERATE INTENSIVE TOTAL ALLUVIUM 35 19 12 66

ANTLERS SAND 9 2 0 11

BOKCHITO FORMATION (UNDIVIDED) 2 0 0 2

EAGLE FORD FORMATION 10 6 2 18

FLUVATILE TERRACE DEPOSITS 34 18 12 64

GOODLAND LIMESTONE AND WALNUT CLAY (UNDIVIDED) 6 1 1 8

GRAYSON MARL AND MAIN STREET LIMESTONE 2 0 1 3

PALUXY SAND 0 1 2 3

PAWPAW FORMATION 4 0 0 4

SURFICIAL DEPOSITS (UNDIVIDED) 0 1 1 2

TWIN MOUNTAINS FORMATION 1 2 0 3

WOODBINE FORMATION 39 18 11 68

281

Table A5. Raw data for hypothesis testing.

ELM/WEST FORKS OCCUPATION DENSITYSOIL SETTING EPHEMERAL MODERATE INTENSIVE TOTAL BLACKLAND 36 13 11 60

CLAY LOAM 42 19 13 74

CLAYEY BOTTOMLAND 32 16 12 60

CLAYPAN PRAIRIE 50 21 13 84

DEEP REDLAND 5 1 0 6

DEEP SAND 11 8 7 26

ERODED BLACKLAND 15 7 5 27

LOAMY BOTTOMLAND 53 25 15 93

LOAMY PRAIRIE 0 1 0 1

LOAMY SAND 23 12 7 42

LOW STONY HILL 2 0 0 2

REDLAND 4 1 1 6

ROCKY HILL 1 0 0 1

SANDSTONE 3 1 0 4

SANDY LOAM 66 28 15 109

SANDY 17 8 3 28

SHALLOW CLAY 2 2 1 5

SHALLOW 1 0 0 1

STEEP ADOBE 7 1 2 10

TIGHT SANDY LOAM 10 5 2 17

WATER 56 26 14 96

TOTAL SITES 68 28 15 111

ELM/WEST FORKS OCCUPATION DENSITYPHYSIOGRAPHIC REGION EPHEMERAL MODERATE INTENSIVE TOTAL

BLACKLAND PRAIRIE 10 5 1 16

EAST CROSS TIMBERS 39 14 10 63

GRAND PRAIRIE 8 4 2 14

ROLLING PLAINS 1 0 0 1

WEST CROSS TIMBERS 10 5 2 17

RIPARIAN ZONE (ELM FORK) (2KM) 14 4 2 20

RIPARIAN ZONE (WEST FORK) (2KM) 1 1 1 3

RIPARIAN ZONE (DENTON CREEK) (2KM) 3 1 0 4

RIPARIAN ZONE (MINOR TRIBUTARIES) (2KM) 28 8 8 44

EPHEMERAL STREAMS (1KM) 17 8 7 32

ECOTONE (BLACKLAND PRAIRIE/EAST CROSS TIMBERS) (2KM) 4 1 1 6

ECOTONE (EAST CROSS TIMBERS/GRAND PRAIRIE) (2KM) 12 6 3 21

ECOTONE (GRAND PRAIRIE/WEST CROSS TIMBERS) (2KM) 6 1 1 8

TOTAL 68 28 15 111

282

Table A6. Raw data for hypothesis testing.

ELM/WEST FORKS ACTIVITIESGREAT GROUP TOOL MANUFACTURE FOOD PREP HUNTING LITHIC SCATTER ARGIUSTOLLS 3 4 4 4 CALCIUSTOLLS 9 16 15 14 HAPLUDERTS 15 24 19 17 HAPLUSTALFS 18 34 35 38 HAPLUSTEPTS 18 35 32 23 HAPLUSTERTS 17 35 33 29 HAPLUSTOLLS 21 40 34 21 PALEUSTALFS 26 45 44 40 PALEUSTOLLS 7 7 3 11 RHODUSTALFS 1 1 0 1 USTIFLUVENTS 15 20 24 16 TOPOGRAPHIC SETTING TOOL MANUFACTURE FOOD PREP HUNTING LITHIC SCATTER BOTTOMLANDS 13 17 21 12 FLOODPLAINS 23 43 40 27 LOW HILLS AND RIDGES 16 29 31 30 OLD TERRACES/VALLEY FILLS 1 2 2 1 PLAINS 3 3 5 3 PLEISTOCENE AGE TERRACES 12 27 28 19 RIDGES 4 6 6 5 STREAM DIVIDES 3 0 1 8 STREAM TERRACES 20 36 36 33 TERRACES 14 24 25 31 UPLAND RIDGES 2 6 5 8 UPLANDS 26 45 44 39 TERRACES, FOOTSLOPES, FANS 14 28 27 19 GEOLOGICAL SETTING TOOL MANUFACTURE FOOD PREP HUNTING LITHIC SCATTER ALLUVIUM 21 31 31 18 ANTLERS SAND 0 4 2 5 BOKCHITO FORMATION 0 0 0 2 EAGLE FORD FORMATION 2 7 8 6 FLUVATILE TERRACE DEPOSITS 21 32 32 19 GOODLAND LIMESTONE AND WALNUT CLAY 1 4 2 3 GRAYSON MARL AND MAIN STREET LIMESTONE 0 1 1 2 PALUXY SAND 2 2 3 0 PAWPAW FORMATION 0 0 0 4 SURFICIAL DEPOSITS 1 1 2 0 TWIN MOUNTAINS FORMATION 1 1 2 0 WOODBINE FORMATION 14 26 30 23

283

Table A7. Raw data for hypothesis testing.

ELM/WEST FORKS ACTIVITIES

SOIL SETTINGTOOL MANUFACTURE

FOOD PREPARATION

HUNTING ACTIVITIES

LITHIC SCATTER

BLACKLAND 14 25 26 27 CLAY LOAM 19 36 32 24 CLAYEY BOTTOMLAND 21 29 29 17 CLAYPAN PRAIRIE 21 36 35 36 DEEP REDLAND 1 0 0 3 DEEP SAND 7 12 16 7 ERODED BLACKLAND 3 12 13 10 LOAMY BOTTOMLAND 23 43 40 27 LOAMY PRAIRIE 1 1 1 0 LOAMY SAND 7 19 21 11 LOW STONY HILL 1 1 0 0 REDLAND 2 3 2 2 ROCKY HILL 1 0 0 0 SANDSTONE 0 1 1 2 SANDY LOAM 26 45 44 39 SANDY 4 9 11 11 SHALLOW CLAY 1 2 3 1 STEEP ADOBE 2 6 3 2 TIGHT SANDY LOAM 4 6 7 5 ELM/WEST FORKS ACTIVITIES

PHYSIOGRAPHIC REGIONTOOL MANUFACTURE

FOOD PREPARATION

HUNTING ACTIVITIES

LITHIC SCATTER

BLACKLAND PRAIRIE 7 11 6 5 EAST CROSS TIMBERS 11 22 25 27 GRAND PRAIRIE 4 6 6 3 WEST CROSS TIMBERS 3 7 7 5 RIPARIAN ZONE (ELM FORK) (2KM) 5 8 5 4 RIPARIAN ZONE (WEST FORK) (2KM) 2 1 2 0 RIPARIAN ZONE (DENTON CREEK) (2KM) 0 1 1 2 RIPARIAN ZONE (MINOR TRIBUTARIES) (2KM) 12 17 18 19 EPHEMERAL STREAMS (1KM) 8 14 15 10 ECOTONE (BLACKLAND PRAIRIE/EAST CROSS TIMBERS) (2KM) 3 2 2 4 ECOTONE (EAST CROSS TIMBERS/GRAND PRAIRIE) (2KM) 6 9 8 6 ECOTONE (GRAND PRAIRIE/WEST CROSS TIMBERS) (2KM) 1 5 2 2

284

Table A8. Raw data for hypothesis testing.

ELM/WEST FORKS OCCUPATION FREQUENCY CHANGE THROUGH TIMEOCCUPATION DENSITY SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII EPHEMERAL 64 4 0 0 51 4 2 0 5 10 0 0 MODERATE 6 15 2 4 6 15 8 3 16 11 2 9 INTENSIVE 1 1 5 8 3 7 9 5 13 14 10 8 ACTIVITIES TOOL MANUFACTURE 11 5 5 5 12 7 9 3 12 12 7 7 CORES HAMMERSTONES PREFORMS BIFACES DEBITAGE FOOD PREPARATION 17 14 5 10 18 15 14 6 20 24 12 14 TOOLS GROUNDSTONE CERAMICS ANIMAL BONES FCR HUNTING ACTIVITIES 8 16 8 12 9 21 17 8 30 26 12 17 DIAGNOSTIC POINTS LITHIC SCATTER 39 1 0 0 31 2 0 0 4 4 0 0 ISOLATED LITHIC DEBRIS

285

Table A9. Economic values: Vegetation scores for soil settings. Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other Boxelder ACERACEAE Acer various 0% 0% 0% 0% 100% Yarrow ASTERACEAE Achillea millefolium L. 50% 50% 0% 0% 0% Canada garlic LILIACEAE Allium canadense 0% 0% 0% 0% 100% Drummond onion LILIACEAE Allium mutabile 0% 0% 0% 0% 100% Wild onion LILIACEAE Allium sp. various 0% 0% 0% 0% 100% Amaranth AMARANTHACEAE Amaranthus spinosis L. 100% 0% 0% 0% 0% Red-Root Rigweed AMARANTHACEAE Amaranthus retroflexus L. 100% 0% 0% 0% 0% Western ragweed ASTERACEAE Ambrosia various 50% 50% 0% 0% 0% Bluestem, big POACEAE Andropogon gerardii 0% 0% 0% 50% 50% Bluestem, sand POACEAE Andropogon various 0% 0% 0% 50% 50% Potato bean FABACEAE Apios americana 100% 0% 0% 0% 0% Threeawn, perennial POACEAE Aristida various 50% 0% 0% 50% 0% Threeawn, purple POACEAE Aristida various 0% 0% 0% 100% 0% Threeawn, Wright POACEAE Aristida various 0% 0% 0% 100% 0% Sagewort ASTERACEAE Artemisia various 50% 50% 0% 0% 0% Butterfly weed ASCLEPIADACEAE Asclepias tuberosa 10% 90% 0% 0% 0% Milkweed ASCLEPIADACEAE Asclepias verticillata L. 10% 90% 0% 0% 0% Whorled milkweed ASCLEPIADACEAE Asclepias verticillata 10% 90% 0% 0% 0% Green antelope horn ASCLEPIADACEAE Asclepias viridis 10% 90% 0% 0% 0% Heath aster ASTERACEAE Aster various 0% 0% 0% 0% 0% Bluestem, cane POACEAE Bothriochloa various 0% 0% 0% 100% 0% Bluestem, silver POACEAE Bothriochloa various 0% 0% 0% 100% 0% Grama, blue POACEAE Bouteloua various 0% 0% 0% 100% 0% Grama, hairy POACEAE Bouteloua various 0% 0% 0% 100% 0% Grama, sideoats POACEAE Bouteloua various 0% 0% 0% 100% 0% Grama, tall POACEAE Bouteloua various 0% 0% 0% 100% 0% Buffalograss POACEAE Buchloe various 0% 0% 0% 100% 0% W. bumelia SAPOTACEAE Bumelia lanuginosa 100% 0% 0% 0% 0% American Beauty Berry VERBENACEAE Callicarpa americana 0% 50% 50% 0% 0% Winecup MALVACEAE Callirhoe various 0% 0% 0% 100% 0% Eveningprimrose ONAGRACEAE Calylophus various 100% 0% 0% 0% 0% Halfshrub sundrop ONAGRACEAE Calylophus various 0% 0% 0% 0% 0% Thistle ASTERACEAE Carduus austrinus 10% 90% 0% 0% 0% Pecan FABACEAE Carya illinoesis 100% 0% 0% 0% 0% Coffee senna FABACEAE Cassia fasciculata 100% 0% 0% 0% 0% Showy partridge pea FABACEAE Cassia occidentalis 100% 0% 0% 0% 0% Scarlet Paint Brush SCROPHULARIACEAE Castilleja Englemann 0% 100% 0% 0% 0% Hackberry ULMACEAE Celtis occidentalis 25% 0% 0% 0% 75% Sugar hackberry ULMACEAE Celtis laevigata 25% 0% 0% 0% 75% Grass Bar POACEAE 0% 50% 0% Cenchrus incertus 50% 0% Basket flower ASTERACEAE Centaurea americana 0% 0% 100% 0% 0% Redbud FABACEAE Cercis canadensis 25% 25% 0% 0% 50% Prairie senna FABACEAE 100% Chamaecrista fasciculata 0% 0% 0% 0% Partridge Pea FABACEAE Chamaecrista Michaux 0% 0% 100% 0% 0% Windmillgrass POACEAE Chloris various 50% 0% 0% 50% 0% Yellow spine thistle ASTERACEAE Cirsum ochrocentrum 50% 50% 0% 0% 0% Wavyleaf thistle ASTERACEAE Cirsum undulatum 50% 50% 0% 0% 0% Virginia springbeauty PORTULACCACEAE Claytonia virginica 100% 0% 0% 0% 0% Bullnettle EUPHORBIACEAE 0% Cnidoscolus texanus 100% 0% 0% 0% Carolina jointtail 0% 0% POACEAE Coelorachis various 0% 0% 100% Dayflower COMMELINACEAE 0% 0% Commelina various 100% 0% 0% Hawthorn ROSACEAE 0% Crataegus various 100% 0% 0% 0% Haw ROSACEAE sp. various 100% 0% 0% 0% 0% Croton EUPHORBIACEAE Croton various 0% 100% 0% 0% 0% Yellow Nutsedge CYPERACEAE Cyperus esculentus 100% 0% 0% 0% 0% Dalea FABACEAE Dalea various 100% 0% 0% 0% 0% Prairie-clover FABACEAE Dalea purpurea 100% 0% 0% 0% 0% White prairie clover FABACEAE Dalea candide 75% 25% 100% 0% 0%

286

Table A10. Economic values: Vegetation scores for soil settings.

Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other

Bundleflower FABACEAE Desmanthus various 0% 0% 0% 100% 0%

Tickclover FABACEAE Desmodium various 0% 0% 0% 100% 0%

Arizona cottontop POACEAE Digitaria californica 50% 0% 0% 50% 0%

Fall witchgrass various POACEAE Digitaria 0% 0% 0% 100% 0%

Wood Sorrel OXALIDACEAE Jacquin 0% dillenii 0% 100% 0% 0%

Common persimmon EBENACEAE virginiana Diospyros 100% 0% 0% 0% 0%

Persimmon Diospyros virginiana L. 0% EBENACEAE 50% 50% 0% 0%

Blacksamson ASTERACEAE Echinacea angustifolia 0% 100% 0% 0% 0%

Coneflower ASTERACEAE Echinacea various 0% 100% 0% 0% 0%

Wildrye, Canada POACEAE Elymus various 50% 0% 0% 50% 0%

Wildrye, Virginia POACEAE Elymus various 0% 50% 0% 0% 50%

Engelmann-daisy ASTERACEAE Engelmannia various 100% 0% 0% 0% 0%

Ephedra EPHEDRACEAE Ephedra various 0% 0% 100% 0% 0%

Lovegrass, plains POACEAE Eragrostis various 0% 50% 0% 0% 50%

Lovegrass, purple POACEAE Eragrostis various 0% 50% 0% 0% 50%

Lovegrass, red POACEAE Eragrostis various 50% 0% 0% 50% 0%

Lovegrass, sand POACEAE Eragrostis various 0% 50% 50% 0% 0%

Texas cupgrass POACEAE Eriochloa various 0% 50% 0% 50% 0%

L. wild buckwheat POLYGONACEAE Eriogonum longifolium 100% 0% 0% 0% 0%

Buckwheat POLYGONACEAE Eriogunum various 100% 0% 0% 0% 0%

White ash OLEACEAE Fraxinus americana 50% 0% 0% 0% 50%

Snakecotton AMARANTHACEAE Frielichia various 0% 100% 0% 0% 0%

Downy milkpea FABACEAE Galactia volubilis 75% 25% 0% 100% 0%

Honey locust FABACEAE Gleditsia tricanthos 40% 0% 10% 0% 50%

C. sunflower ASTERACEAE Helianthus annus 0% 50% 50% 0% 0%

M. sunflower ASTERACEAE Helianthus various 50% 50% 0% 0% 0%

Sunflower ASTERACEAE Helianthus anmus l. 50% 0% 50% 0% 0%

Curlymesquite POACEAE Hilaria various 50% 0% 0% 50% 0%

Western Indigo FABACEAE Indigofera various 0% 0% 0% 100% 0%

Black Walnut FABACEAE Juglans nigra L. 100% 0% 0% 0% 0%

Rush FABACEAE Juncus various 100% 0% 0% 0% 0%

Juniper CUPRESSACEAE Juniperus various 75% 0% 0% 25% 0%

Red Cedar PINACEAE Juniperus virginiana 50% 0% 25% 0% 25%

Trailing ratany FABACEAE Krameria various 100% 0% 0% 0% 0%

G. sprangletop POACEAE Leptochloa various 0% 50% 0% 0% 50%

Lespedeza FABACEAE Lespedeza various 0% 0% 50% 0% 50%

Honeysuckle CAPRIFOLIACEAE Lonicera various 0% 0% 0% 0% 0%

T.Honeysuckle CAPRIFOLIACEAE Lonicera sempervirens 0% 100% 0% 0% 0%

Bluebonnet FABACEAE Lupinus texensis 0% 0% 0% 100% 0%

Mint FABACEAE Mentha sp. various 0% 100% 0% 0% 0%

Catclaw FABACEAE Mimosa 50% various 50% 0% 0% 0%

Sensitivebriar FABACEAE Mimosa various 0% 0% 0% 50% 50%

Horsemint LAMIACEAE 100% Monarda citriodora C. 0% 0% 0% 0%

Red mulberry MORACEAE rubra 0% Morus 75% 25% 0% 0%

Seep muhly POACEAE Muhlenbergia various 0% 50% 0% 0% 50%

T. wintergrass POACEAE Nassella various 50% 0% 0% 50% 0%

Water lily NYMPHAEACEAE Nelubo 0% 0% 0% 0% lutea 100%

287

Table A11. Economic values: Vegetation scores for soil settings. Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other Common prickly pear Opuntia compressa 0% 0% 0% CACTACEAE 75% 25% Prickly Pear CACTACEAE Opuntia phaeacantha E. 50% 50% 0% 0% 0% Agrito OXALIDACEAE corniculata 0% 75% 0% 0% Oxalis 25% Panicum, low POACEAE Panicum various 50% 0% 0% 50% 0% Panicum, Scribner POACEAE Panicum various 0% 0% 50% 0% 50% Switchgrass POACEAE Panicum various 50% 0% 0% 50% 0% Vine-mesquite POACEAE Panicum various 0% 0% 0% 100% 0% Western wheatgrass POACEAE Pascopyrum 0% 0% various 50% 0% 50% Paspalum, Florida POACEAE Paspalum 50% various 50% 0% 0% 0% Paspalum, fringeleaf POACEAE Paspalum various 50% 0% 0% 50% 0% Scurfpea FABACEAE Pediomelum various 100% 0% 0% 0% 0% Penstemon SCROPHULARIACEAE Penstemon various 0% 0% 0% 0% 0% Purple groundcherry SOLANACEAE lobata 0% 0% Physalis 75% 0% 25% Common pokeberry PHYTOLACCACEAE Phytolacca various 0% 50% 0% 25% 25% Buckthorn PLANTAGINACEAE Plantago various 0% 0% 0% 0% 0% Plantain PLANTAGINACEAE Plantago sp. various 50% 50% 0% 0% 0% Texas bluegrass POACEAE Poa various 50% 0% 0% 50% 0% Milkwort POLYGALACEAE Polygala various 75% 25% 0% 0% 0% Cottonwood SALICACEAE Populus D. 0% 50% 0% 0% 50% Common mesquite FABACEAE Prosopis juliflora 70% 0% 0% 0% 30% Honey mesquite FABACEAE Prosopis glandulosa 70% 0% 0% 0% 30% Mesquite Prosopis 0% 0% 30% FABACEAE glandulosa C 70% 0% American plum ROSACEAE Prunus 100% 0% 0% 0% americana 0% Chickasaw plum ROSACEAE Prunus angustifolia 80% 10% 0% 0% 10% Common chokecherry ROSACEAE Prunus virginiana 100% 0% 0% 0% 0% Oklahoma plum ROSACEAE Prunus gracilis 100% 0% 0% 0% 0% Mexican plum ROSACEAE Prunus 10% 0% 10% mexicana 80% 0% Wild plum ROSACEAE Prunus sp. various 100% 0% 0% 0% 0% Indian breadroot FABACEAE Psoralea hypogeae 100% 0% 0% 0% 0% Wild alfafa FABACEAE Psoralidium various 50% 0% 0% 50% 0% Caroline false-dandelion ASTERACEAE Pyrrhopappus carolinianus 50% 50% 0% 0% 0% Oak, bur FABACEAE Quercus macrocarpa 100% 0% 0% 0% 0% Oak, blackjack FABACEAE Quercus marilandica 100% 0% 0% 0% 0% Oak, willow FAGACEAE Quercus phellos L. 0% 0% 0% 0% 100% Shumard oak FABACEAE Quercus shumardii 100% 0% 0% 0% 0% Oak FABACEAE Quercus sp. various 100% 0% 0% 0% 0% Oak, post FABACEAE Quercus stellata 100% 0% 0% 0% 0% Oak, live FAGACEAE Quercus virginiana Mill 0% 25% 0% 0% 75% Skink bush aromatica 0% ANACARDIACEAE Rhus 10% 75% 0% 15% Flame leaf sumac ANACARDIACEAE 0% Rhus copallina 50% 25% 0% 25% Smooth sumac ANACARDIACEAE Rhus glabra 25% 50% 25% 0% 0% Sumac 25% 0% ANACARDIACEAE Rhus spp. various 50% 0% 25% Poison Ivy ANACARDIACEAE Rhus toxicodendron L. 0% 100% 0% 0% 0% Snoutbean 0% 50% FABACEAE Rhynchosia various 50% 0% 0% Clove current SAXIFRAGACEAE 0% 0% 0% Ribes odoratum 100% 0% Black Locust FABACEAE Robinia pseudo-acacia L. 0% 0% 25% 0% 75% Black raspberry ROSACEAE Rubus occidentialis 100% 0% 0% 0% 0% Dewberry ROSACEAE Rubus aboriginum 100% 0% 0% 0% 0% Red raspberry ROSACEAE Rubus idaeaus 100% 0% 0% 0% 0% Louisiana blackberry ROSACEAE Rubus lousianus 100% 0% 0% 0% 0% Southern Dewberry ROSACEAE Rubus triviklis M. 100% 0% 0% 0% 0% Ruella ACANTHACEAE Ruellia various 0% 0% 0% 0% 100% Arrowhead ALISMATACEAE Sagittaria latiforlia 100% 0% 0% 0% 0% Black Willow SALICACEAE Salix nigra Marsh. 0% 100% 0% 0% 0% Sage FABACEAE Salvia various 100% 0% 0% 0% 0% Bluestem, little 50% 0% POACEAE Schizachyrium various 50% 0% 0% Soft stem bulrush CYPERACEAE Scirpus validus 100% 0% 0% 0% 0% Skullcap various 100% 0% 0% FABACEAE Scutellaria 0% 0% Knottroot bristlegrass POACEAE various Setaria 50% 0% 0% 50% 0% Sida MALVACEAE 100% 0% 0% 0% 0% Sida various

288

Table A12. Economic values: Vegetation scores for soil settings. Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other Bumelia Sideroxylon 0% 0% 0% 0% SAPOTACEAE various 0% Compassplant ASTERACEAE Silphium laciniatum 0% 100% 0% 0% 0% Bushsunflower ASTERACEAE 0% 100% 0% 0% 0% Simsia various Greenbriar SMILACACEAE Smilax various 100% 0% 0% 0% 0% Saw greenbriar SMILACACEAE Smilax bona-nox 75% 0% 0% 0% 25% Indiangrass POACEAE Sorghastrum various 50% 0% 0% 50% 0% Dropseed, meadow POACEAE Sporobolus various 50% 0% 0% 50% 0% Dropseed, sand POACEAE Sporobolus various 50% 0% 0% 50% 0% Dropseed, tall POACEAE Sporobolus various 50% 0% 0% 50% 0% Falsegaura ONAGRACEAE various Steinosiphon 0% 0% 0% 0% 0% Queensdelight EUPHORBIACEAE Stillingia various 0% 0% 0% 0% 0% Wildbean FABACEAE Strophostyles various 80% 10% 0% 10% 0% Trailing wild bean FABACEAE Strophostyles helvola 80% 10% 0% 10% 0% Coralberry CAPRIFOLIACEAE Symphoricarpos various 0% 0% 100% 0% 0% Tridens, purpletop POACEAE Tridens various 50% 0% 0% 50% 0% Tridens, rough POACEAE Tridens various 50% 0% 0% 50% 0% Tridens, slim POACEAE Tridens various 50% 0% 0% 0% 50% Tridens, white 50% 0% 0% POACEAE Tridens various 0% 50% Clover FABACEAE Trifolium repens L. 25% 25% 0% 50% 0% Eastern gamagrass POACEAE Tripsacum various 50% 0% 0% 50% 0% Cattail TYPHACEAE Typha latifolia 90% 0% 10% 0% 0% Cedar Elm ULMACEAE Ulmus crassifolia Nutt 0% 0% 0% 0% 100% Elm, American ULMACEAE Ulmus americana L. 25% 0% 0% 0% 75% Elm, cedar ULMACEAE 0% 0% 0% 0% 100% Ulmus various Elm, winged ULMACEAE Ulmus various 0% 0% 0% 0% 100% Slippery elm ULMACEAE Ulmus rubra 25% 0% 0% 0% 75% Blue Vervain VERBENACEAE 0% 100% 0% 0% 0% Verbena halei Verbena VERBENACEAE Verbena 0% 0% various 100% 0% 0% Ironweed ASTERACEAE Verononia various 0% 0% 0% 0% 0% Haw ROSACEAE Viburnum prunifolium 0% 0% 75% 25% 0% Vetch 0% 0% 0% 0% FABACEAE Vicia various 100% Grape VITACEAE Vitis spp. various 50% 50% 0% 0% 0% Soap weed AGAVACEAE Yucca glauca 50% 50% 0% 0% 0% Yucca AGAVACEAE Yucca louisianensis T. 0% 0% 50% 25% 25% Prickly Ash RUTACEAE Zanthoxylum clavaherculis L. 0% 100% 0% 0% 0% Indian Corn FABACEAE 100% 0% 0% 0% 0% Zea mays various Lotebush RHAMNACEAE Ziziphus various 100% 0% 0% 0% 0% Dropseed, hairy POACEAE various 50% 0% 0% 0% 50% Grain Sorghum POACEAE 50% 0% 0% 50% various 0% L. hackberry ULMACEAE various 0% 0% 0% 0% 100% Oak, shin FAGACEAE various 0% 0% 100% 0% 0% Oak, Texas FAGACEAE various 0% 50% 50% 0% 0% Peanuts FABACEAE various 100% 0% 0% 0% 0% Roughleaf dogwood CORNACEAE 0% 0% 0% 100% various 0% Sedge CYPERACEAE 100% 0% 0% 0% various 0% Stemless actinea POACEAE various 50% 0% 50% 0% 0% Texas ash OLEACEAE various 0% 0% 0% 0% 100% Texas sophora FABACEAE various 0% 0% 0% 0% 0% Western soapberry SAPINDACEAE 0% 0% 100% 0% 0% various Wheat POACEAE various 100% 0% 0% 0% 0%

289

Table A13. Economic values of site catchment areas in the Upper Trinity Basin.

ECONOMIC VALUE OF TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES EDIBILITY VALUE MEDICINAL VALUE SUPPLEMENTAL VALUE ECONOMIC VALUE TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean St Dev Var/Mean Sum Mean St Dev Sum Var/Mean Mean St Dev Var/MeanARCHAIC 0.1800 0.0450 0.0107 0.0026 0.1576 0.0394 0.0116 0.0034 0.1709 0.0427 0.0074 0.0013 0.1689 0.0422 0.0095 0.00214EARLY ARCHAIC 0.1324 0.0441 0.0122 0.1257 0.0419 0.0073 0.0013 0.0034 0.1176 0.0392 0.0133 0.0045 0.1242 0.0414 0.0107 0.00276MIDDLE ARCHAIC 0.1902 0.0475 0.0121 0.0081 0.0031 0.1743 0.0436 0.0138 0.0044 0.1790 0.0448 0.0014 0.1788 0.0447 0.0109 0.00265 LATE ARCHAIC 0.0108 0.0025 0.2133 0.0427 0.0125 0. 0.0443 0.0098 0.00217 0.2360 0.0472 0037 0.2196 0.0439 0.0740 0.1247 0.2213LATE PREHISTORIC 0.1596 0.0399 0.0129 0.1541 0.1138 0.34492 0.0041 0.1397 0.0349 0.0137 0.0054 0.0385 0.0087 0.0019 0.1503 0.0376 LATE PREHISTORIC I 0.0568 0.0568 0.0000 0.0000 0.0479 0.0000 0.0551 0.0551 0.0000 0.0479 0.0000 0.0000 0.0524 0.0524 0.0000 0.0000 LATE PREHISTORIC II 0.1543 0.0514 0.0039 0.1450 0.0019 0.0001 0.1444 0.0481 0.0032 0.0003 0.1360 0.0453 0.0069 0.0011 0.0483 0.00021 PREHISTORIC 0.5838 0.0486 0.0081 0. 0.0089 0.0017 0.5498 0.0458 0.0078 0014 0.5128 0.0427 0.0101 0.0024 0.5634 0.0469 0.00133 TRINITY ELM FORK ARCHAIC 0.0120 0.0033 1.1491 0.0522 0.0187 0.0067 1.2124 0.0551 0.0135 0.0033 0.9599 0.0436 1.1196 0.0509 0.0137 0.00369EARLY ARCHAIC 0.7376 0.0461 0.0128 0.0036 0.8137 0.0509 0.0039 0.0003 0.6638 0.0414 0.0047 0.0005 0.7446 0.0465 0.0045 0.00044MIDDLE ARCHAIC 0.0018 0.2593 0.0432 0.0059 0.0008 0.0444 0.0120 0.2879 0.0479 0.0001 0.0046 0.0005 0.2522 0.0420 0.2665 0.00324 LATE ARCHAIC 0.0026 1.2129 0.0433 0.0118 1.5218 0.0544 0.0120 0.0032 1.3979 0.0499 0.0183 0.0067 1.3971 0.0499 0.0123 0.00303LATE PREHISTORIC 1.6021 0.0517 0.0050 0.0005 1.3266 0.0428 0.0056 0.0007 1.4584 0.0470 0.0139 0.0041 1.4785 0.0477 0.0061 0.00078 LATE PREHISTORIC I 0.5468 0.0497 0.0032 0.4859 0.0002 0.4871 0.0443 0.0059 0.0008 0.0442 0.0110 0.0028 0.5078 0.0462 0.0046 0.00046 LATE PREHISTORIC II 0.6994 0.0119 0.0500 0.0028 0.0002 0.5850 0.0418 0.0045 0.0005 0.6075 0.0434 0.0033 0.6396 0.0457 0.0044 0.00042 PREHISTORIC 2.5729 0.0547 0.0091 0.0015 1.9450 0.0414 0.0098 0.0023 2.2058 0.0469 0.0147 0.0046 2.3181 0.0493 0.0081 0.00133

ECONOMIC VALUE OF TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES EDIBILITY VALUE MEDICINAL VALUE SUPPLEMENTAL VALUE ECONOMIC VALUE TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean St Dev Var/Mean Sum Mean St Dev Var/Mean Sum Mean St Dev Var/MeanSINGLE 0.5800 0 0.0486 0.0080 0.0013 0.512 0.0420 0.0100 0.0024 0.5800 0.0486 0.0080 0.0013 0.5490 0.0450 0.0080 0.00142DUAL 0.0079 0.1532 0.0511 0.0049 0.0005 0.1367 0.0456 0.0014 0.1448 0.0483 0.0055 0.0006 0.1442 0.0481 0.0050 0.00052TRI 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000MULTI 0.1324 0.0441 0.0122 0.0034 0.1176 0.0392 0.0133 0.0045 0.1257 0.0419 0.0073 0.0013 0.1242 0.0414 0.0107 0.00277 TRINITY ELM FORK SINGLE 3.1760 0.0530 0.0080 0.0012 2.4400 0.0410 0.0080 0.0016 2.8000 0.0470 0.0140 0.0042 2.8800 0.0480 0.0070 0.00102DUAL 0.9838 0.0578 0.0143 0.0035 0.7718 0.0454 0.0141 0.0044 0.9115 0.0536 0.0227 0.0096 0.9043 0.0532 0.0150 0.00423TRI 0.4098 0.0512 0.0042 0.0003 0.3451 0.0431 0.0062 0.0009 0.3909 0.0489 0.0063 0.0008 0.3270 0.0478 0.0036 0.00027MULTI 0.4500 0.0500 0.0024 0.0001 0.3772 0.0419 0.0052 0.0006 0.3889 0.0432 0.0115 0.0031 0.4110 0.0457 0.0044 0.00042

290

Table A14. Vegetation potentials of site catchment areas in the Upper Trinity Basin.

POTENTIAL VEGETATION TYPE WITHIN TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES GRAINS AND SEED CROPS GRASSES AND LEGUMES WILD HERBACEOUS PLANTS SHRUBS

TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean

St Dev Var/Mean Sum Mean

St Dev Var/Mean Sum Mean

St Dev Var/Mean

ARCHAIC 0.918 0.0720.230 0.0224 1.307 0.327 0.129 0.0505 1.551 0.388 0.117 0.0350 1.688 0.422 0.117 0.03237EARLY ARCHAIC 0.678 0.226 0.077 0.0260 0.913 0.304 0.141 0.0657 1.033 0.344 0.107 0.0331 1.133 0.378 0.105 0.02933MIDDLE ARCHAIC 1.014 0.254 0.082 0.0264 1.277 0.319 0.125 0.0491 1.360 0.340 0.093 0.0253 1.546 0.386 0.092 0.02205LATE ARCHAIC 1.233 0.247 0.074 0.0225 1.667 0.333 0.116 0.0400 1.853 0.371 0.103 0.0287 2.076 0.415 0.101 0.02438LATE PREHISTORIC 0.781 0.195 0.085 0.115 0.0437 1.0.0374 1.119 0.280 0.130 0.0601 1.218 0.304 326 0.332 0.121 0.04439 LATE PREHISTORIC I 0.276 0.276 0.000 0.0000 0.408 0.408 0.000 0.0000 0.397 0.397 0.000 0.0000 0.410 0.410 0.000 0.000LATE PREHISTORIC II 0.866 0.289 0.013 0.0005 1.213 0.404 0.003 0.0000 1.303 0.434 0.029 0.0019 1.410 0.470 0.044 0.00403PREHISTORIC 0.407 3.248 0.271 0.075 0.0206 4.584 0.324 0.104 0.0331 4.882 0.096 0.0224 5.231 0.436 0.076 0.01332 TRINITY ELM FORK ARCHAIC 6.093 0.277 0.086 0.0266 0.322 7.092 0.096 0.0286 8.352 0.380 0.104 0.0284 9.094 0.413 0.130 0.04089EARLY ARCHAIC 4.581 0.286 0.094 0.0306 5.263 0.329 0.095 0.0274 6.000 0.375 0.101 0.0271 5.919 0.370 0.107 0.03078MIDDLE ARCHAIC 1.803 0.300 0.063 0.0130 1.953 0.325 0.064 0.0124 2.180 0.363 0.059 0.0096 2.008 0.334 0.058 0.01007LATE ARCHAIC 7.324 0.264 0.091 0.0316 8.454 0.389 0.302 0.099 0.0325 10.337 0.369 0.109 0.0321 10.887 0.133 0.04543 LATE PREHISTORIC 7.483 0.241 0.084 0.02519 0.0290 8.539 0.275 0.083 0.0253 10.730 0.346 0.075 0.0163 11.061 0.357 0.095LATE PREHISTORIC I 2.966 0.270 0.772 2.2102 3.289 0.299 0.068 0.0154 3.818 0.347 0.072 0.0149 3.783 0.344 0.082 0.01931LATE PREHISTORIC II 0.087 3.548 0.253 0.0299 4.028 0.288 0.092 0.0293 4.525 0.347 0.088 0.0224 4.937 0.353 0.101 0.02879PREHISTORIC 1 0.0279 0.102 11.901 0.253 0.102 0.0409 3.721 0.292 0.122 0.0509 14.876 0.317 0.094 16.133 0.343 0.03049

POTENTIAL VEGETATION TYPE WITHIN TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES GRAINS AND SEED CROPS GRASSES AND LEGUMES WILD HERBACEOUS SHRUBS

TRINITY WEST FORK Mean Dev SumSt

Var/Mean Sum MeanSt Dev Var/Mean Sum Mean

St Dev Var/Mean Sum Mean

St Dev Var/Mean

SINGLE 4.689 0.0820.390 0.0172 0.269 3.232 0.077 0.0221 4.872 0.406 0.097 0.0232 5.188 0.432 0.085 0.01672 DUAL 0.00558 1.159 0.0170.386 0.0007 0.861 0.287 0.049 0.0084 1.286 0.429 0.073 0.0124 1.455 0.485 0.052TRI 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000MULTI 0.913 0.304 0.141 0 0.344 0.107 0.0331 1. .0657 678 0.226 0. 0.077 0.0260 1.033 133 0.378 0.105 0.02929 TRINITY ELM FORK SINGLE 16.888 0.286 0.105 0.251 0.104 0.0385 14.838 0.096 0.0367 19.522 0.330 0.098 0.0291 20.835 0.353 0.03064DUAL 0.272 0.163 5.571 0.1400.327 0.0599 4.626 0.113 0.0469 6.234 0.366 0.133 0.0483 6.741 0.396 0.06709

TRI 0.0055 2.416 0.0660.302 0.0144 2.228 0.279 0.070 0.0176 2.943 0.368 0.045 3.138 0.392 0.061 0.00949MU 3. 0.0190 LTI 2.633 0.293 0.073 0.0182 0.255 2.300 0.079 0.0245 107 0.345 0.081 3.107 0.345 0.088 0.02245

291

Table A15. Wildlife potentials of site catchment areas in the Upper Trinity Basin.

WILDLIFE POTENTIAL WITHIN TRINITY RIVER WEST FORK AND EAST FORK SITE CATCHMENTS BY TIME PERIOD ARCHAEOLOGICAL SITES OPENLAND WILDLIFE RANGELAND WILDLIFE WETLAND WILDLIFE WILDLIFE

TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean

St Dev Var/Mean Sum Mean

St Dev Var/Mean Sum Mean

St Dev Var/Mean

ARCHAIC 0.333 0.0378 1.332 0.112 1.483 0.371 0.105 0.0298 0.110 0.028 0.007 0.0019 2.926 0.732 0.224 0.06856 EARLY ARCHAIC 1.034 0.344 0.108 0.679 0.924 0.308 0.119 0.0458 0.0339 0.078 0.026 0.008 0.0022 2.036 0.234 0.08069MIDDLE ARCHAIC 0.095 1.307 0.325 0.107 0.0353 1.412 0.353 0.0257 0.116 0.029 0.008 0.0025 2.827 0.707 0.209 0.06159 LATE ARCHAIC 1.691 0.332 0.099 0.0297 0.05193 1.843 0.369 0.091 0.0224 0.146 0.029 0.076 0.1968 3.680 0.736 0.196 LATE PREHISTORIC 0.0446 0.241 1.074 0.269 0.124 0.0568 1.220 0.305 0.117 0.119 0.030 0.009 0.0030 2.413 0.603 0.09645LATE PREHISTORIC I 0.000 0.0000 0.032 0.380 0.380 0.000 0.0000 0.383 0.383 0.032 0.000 0.0000 0.800 0.800 0.000 0.000LATE PREHISTORIC II 0.095 0.0000 1.205 0.402 0.017 0.0008 1.307 0.436 0.034 0.0027 0.032 0.001 2.607 0.869 0.051 0.00299PREHISTORIC 4.290 0.357 0.091 0.0017 9.515 0.793 0.0231 4.841 0.403 0.086 0.0182 0.385 0.032 0.007 0.172 0.03709 TRINITY ELM FORK ARCHAIC 7.099 0.323 0.079 0.0192 8.007 0.364 0.105 0.0304 1.385 0.063 0.031 0.0150 0.179 16.491 0.750 0.04274EARLY ARCHAIC 5.233 0.327 0.094 0.0272 5.642 0.353 0.098 0.0272 1.106 0.069 0.033 0.0158 11.982 0.748 0.200 0.05332 MIDDLE ARCHAIC 0.0117 0.324 0.053 0.0086 0.536 0.089 0.031 0.0108 4.385 0.738 0.123 0.02064 1.907 0.318 0.061 1.942LATE ARCHAIC 8.593 0.307 0.089 0.0259 9.783 0.349 0.108 0.072 0.728 0.05644 0.0333 2.015 0.035 0.0167 20.391 0.203LATE PREHISTORIC 8.658 0.279 0.073 0.077 0.0212 10.058 0.324 0.0162 2.386 0.077 0.049 0.0307 21.103 0.681 0.147 0.03175LATE PREHISTORIC I 3.270 0.297 0.068 0.0155 3.490 0.317 0.060 0.0113 0.919 0.084 0.041 0.0204 7.679 0.698 0.151 0.0327LATE PREHISTORIC II 4.088 0.292 0.090 0.0279 4.517 0.323 0.078 0.0189 0.949 0.068 0.031 0.0141 9.553 0.682 0.185 0.05001PREHISTORIC 13.418 0.285 0.100 0.0351 14.419 0.307 0.084 0.0229 3.464 0.074 0.045 0.0274 31.301 0.665 0.191 0.05457

WILDLIFE POTENTIAL WITHIN TRINITY RIVER WEST FORK AND EAST FORK SITE CATCHMENTS BY TIME PERIOD ARCHAEOLOGICAL SITES OPENLAND WILDLIFE RANGELAND WILDLIFE WETLAND WILDLIFE WILDLIFE VALUE

TRINITY WEST FORK Sum Sum Mean St Dev Var/Mean Mean

St Dev Var/Mean Sum Mean

St Dev Var/Mean Mean

St Dev Sum Var/Mean

SINGLE 4.299 9. 0.358 0.089 0.0221 4.830 0.402 0.087 0.0188 0.411 0.034 0.005 0.0007 541 0.795 0.165 0.03425DUAL 1.189 0.396 0.019 0.0009 1.277 0.426 0.037 0.0032 0.101 0.034 0.003 0.0003 2.566 0.855 0.053 0.00329 TRI 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000MULTI 0.0458 034 0.345 0.109 0.0343 0.0.924 0.308 0.119 1. 078 0.026 0.008 0.0022 2.036 0.679 0.234 0.08083 TRINITY ELM FORK SINGLE 16.871 .685 0.285 0.093 0.0303 18.998 0.322 0.090 0.0252 4.569 0.077 0.053 0.0365 40.440 0 0.182 0.04836DUAL 5.451 0.321 0.121 0.0456 6.059 0.356 0.135 0.0512 1.119 0.065 0.030 0.0138 12.631 0.743 0.254 0.08683TRI 2.506 0.313 0.050 0.0080 2.729 0.341 0.044 0.0057 0.645 0.081 0.035 0.0151 5.884 0.735 0.101 0.01388MULTI 2.626 0.292 0.074 0.0188 2.876 0.319 0.067 0.0141 0.619 0.068 0.036 0.0191 6.119 0.679 0.159 0.03723

292

GIS GLOSSARY

Buffer: Procedure conducted with ArcMap Geoprocessing tools which generates a circle or polygon at a specified distance around a point.

Dissolve: Procedure conducted with ArcMap Geoprocessing tools that “creates a new data set in which all features in an input layer that have the same value for a specified attribute become a single feature” (Ormsby et al., 2001: 270).

Overlay analysis: A method of analysis that “creates a new data set by superimposing one layer on another. The output data has the attributes of both input layers” (Ormsby et al., 2001: 306).

Raster: An image that uses a “grid structure to store geographic information” (Minami, 2000: 510).

Selection: Selecting data can be conducted in ArcMap by location or attributes. Selecting features by location involves specifying “a selection method, a selection layer, a spatial relationship, a reference layer, and sometimes a distance buffer” (Ormsby et al., 2001: 250). Selecting features by attribute involves specifying a specific attribute (i.e. time period) to create a new data layer. Symbology: A term that refers to “criteria used to determine symbols for the features in a layer” (Minami, 2000: 513). Tabular data: A phrase that refers to “descriptive information that is stored in rows and colums and can be linked to map features” (Minami, 2000: 513).

Clip: Procedure conducted with ArcMap Geoprocessing tools that “trims features in one layer at the boundaries of features in another layer” (Ormsby et al., 2001: 282).

Joining: To associate a non-spatial table with a spatial data layer based on a unique identification code (i.e. soils series). This assigns meaning to a spatial data layer.

Projection: the transformation of descriptive data to spatial location with a geographic coordinate system. Query: A procedure conducted in ArcMap to extract data that involves “a question or request for selecting features. A query often appears in the form of a statement or logical expression. In ArcMap, a query contains a field, an operator, and a value” (Minami, 2000: 510).

293

REFERENCES

Urbanism. edited by Peter Ucko, Ruth Tringham and G.W. Dimbleby. Proceedings

Allen, K.M.S., Green, S.W. and Zubrow, E.B.W. Interpreting Space: GIS and Archaeology,

.

Biesaart, L.A., W.R. Roberson, L.C. Spotts. Prehistoric Archaeological Sites in Texas: A Statistical Overview. Office of the State Archaeologist Special Report 28. Austin: Texas Historical Commission, 1985.

Binford, L.R. “The Archaeology of Place.” Journal of Anthropological Archaeology 1(1)

(1982): 5-31. _____.Constructing Frames of Reference: An Analytical Method for Archaeological

Theory Building Using Hunter-Gatherer and Environmental Data Sets. Berkeley, California; London : UC Press, 2002.

Bousman, B. and L. Verrett. An Archaeological Reconnaissance at Aubrey Reservoir.

Archaeology Research Program, Southern Methodist University, Dallas. Fort Worth: U.S. Army Corps of Engineers, 1973.

Bruseth, James E., Daniel E. McGregor, and William A. Martin. “Hunter-Gatherers of the

Allan, W. “Ecology, Techniques and Settlement Patterns.” In: Man, Settlement and

of a meeting of the Research Seminar in Archaeology and Related Subjects held at the Institute of Archaeology, London University. Cambridge, Massachusetts: Schenkman Publishing Company, 1973.

London: Taylor and Francis, 1990.

Bernick, Kathryn N. A Site Catchment Analysis of the Little Qualicum River Site, DiSc 1, Wet Site on the East Coast of Vancouver Island, B C. Ottawa: National Museums of Canada, 1983.

Brown, K.L. and B.C. Yates. Archaeological Testing of the Lewisville Lake Shoreline, Denton County, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1990.

Brown, K.L. and S.A. Lebo. Archaeological Testing of the Lewisville Lake Shoreline,

Denton County, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1991.

Prairie Margin: Summary of the Prehistoric Archaeological Record.” In: Hunter-Gatherer Adaptations Along the Prairie Margins: Site Excavations and Synthesis of Prehistoric Archaeology, edited by D.E. McGregor and J.E. Bruseth. Dallas: Richland Creek Technical Series, Volume III. Archaeology Research Program, Southern Methodist University, 1987.

294

Butzer, Karl M. Environment and Archeology: An Ecological Approach to Prehistory. Chicago: Aldine-Atherton, 1971.

Christaller, Walter. Central Places in Southern Germany. Translated from Die zentralen

Orte in Süddeutschland by Carlisle W. Baskin. New Jersey: Englewood Cliffs, Prentice-Hall, 1966.

f

Dunnell, Robert C. Systematics in Prehistory. New York: Free Press, 1971. Dyksterhuis, E.J. “The Vegetation of the Fort Worth Prairie.” In: Ecological Monographs

_____.Archaeology as Human Ecology. Cambridge University Press: Cambridge, 1982. Chisholm, Michael. Rural Settlement and Land Use. London: Hutchinson, 1968.

Clarke, David. Models in Archaeology. London: Methuen, 1972. Coe M.D. and K.V. Flannery “Microenvironments and Mesoamerican Prehistory”, In:

Science: New Roads to Yesterday, edited by J.R. Caldwell, New York/London, 1966.

Colinvaux, Paul. Ecology 2. New York: John Wiley & Sons, Inc., 1993. Dawson, G.L. and T.L. Sullivan. Excavations at Lake Lavon: 1969. Dallas: Report

submitted to the National Park Service by the Archaeology Research Program, Department of Anthropology, Southern Methodist University, 1969.

Dennell, R.W. “The Economic Importance of Plant Resources Represented on

Archaeological Sites.” Journal o Archaeological Science 3 (1976): 229-47. Diggs, G.M., B.L. Lipscomb and R.J. O’Kennon. Shinners and Mahler’s Illustrated Flora

of North-Central Texas. Fort Worth: Botanical Research Institute of Texas, 1999. Dillehay, T.D. “Late Quaternary Bison Population Changes on the Southern Plains.”

Plains Anthropologist 19 (65) (1974): 180-196. Duke, James. Eastern/Central Medicinal Plants and Herbs. New York: Houghton Mifflin

Company, 2000.

Durham: Duke University Press, 1946. _____.“The Vegetation of the Western Cross Timbers.” In: Ecological Monographs

Durham: Duke University Press, 1948. Ferring, C.R. The Archaeology and Paleoecology of the Aubrey Clovis Site (41DN479)

Denton County, Texas. Denton: Center for Environmental Archaeology, Department of Geography, University of North Texas, 2001.

295

Ferring and B.C. Yates. An Assessment of the Cultural Resources in the Trinity Basin, Dallas, Tarrant, and Denton Counties, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1986.

_____.Holocene Geoarchaeology and Prehistory of the Ray Roberts Lake Area, North- Central Texas. Denton: Institute of Applied Sciences, University of North Texas, 1997.

_____.Prehistory of Archaeological Sites at Lewisville Lake, North-Central Texas.

and Lateral Watershed, Wise County, Texas. Denton: Institute Environmental

Denton: Institute of Applied Sciences, University of North Texas, 1998 Ferring, C. Reid, Britt Bousman, Barbara Butler, Dan Crouch, Steven Hackenberger,

Steve A. Hall, Roger Hoestenbach, Kevin Leehan, Jim Robertson, and Richard Vernon. An Archaeological Reconnaissance of Fort Sill, Oklahoma. Lawton, Oklahoma: Contributions of the Museum of the Great Plains Number 6, 1978.

Filson, Roger E., and Olin F. McCormick. Archaeological Survey of Portions of Salt Creek

Studies. North Texas State University, 1975. Flannery, Kent V. "The Village and its Catchment Area: Introduction.” In: The Early

Mesoamerican village. edited by K.V. Flannery. New York: Academic Press, 1968.

_____."Empirical determination of Site Catchments in Oazaca and Tehuacan. In: The Early Mesoamerican village. edited by K.V. Flannery. New York: Academic Press, 1976.

Ford, A. and E. Pauls. Soil Survey of Denton County, Texas. : Washington D.C., U.S.

Department of Agriculture, Soil Conservation Service, U.S. Government Printing Office, 1980.

Geier, Clarence. “Some Ecological Concepts and Their Implications for Settlement

Archaeology.” Plains Anthropologist 19(63) (1974): 46-54. Goodchild, Michael F. “Geographic Information Systems and Spatial Analysis in the

Social Sciences” In: Anthropology, Space, and Geographic Information Systems. edited by Mark Aldenderfer and Herbert D.G. Maschner. Oxford University Press: New York , 1996.

Griffiths, J. and J. Bryan. The Climates of Texas Counties. College Station: Natural

Fibers Information Center, the University of Texas at Austin in Cooperation with Office of the State Climatologist, Texas A & M University: 1987.

Haggett, Peter. Locational Analysis in Human Geography. New York: St. Martin's Press,

1966.

296

Hanson, Arthur and Frankie F. Wheeler. Soil Survey, Collin County,Texas. U.S. Department of Agriculture, Soil Conservation Service, U.S. Government Printing Office: Washington D.C., 1969.

Hastorf, Christine. “Changing Resource Use in Subsistence Agricultural Groups of the

t

Higgs, E.S. and C. Vita-Finzi. “Prehistoric Economies: A Territorial Approach.” In: Papers

t

rby David L Clarke. London :Methuen, 1972.

_____.“Prehistoric economic development in sub-alpine Italy.” In: Problems in Economic and Social Archaeology edited by G. de G. Sieveking, I. H. Longworth, and K. E. Wilson. Boulder, Westview Press, 1976.

New York: Academic Press, 1976.

r t

Prehistoric Mimbres River Valley, New Mexico.” In: Modeling Change in Prehistoric Subsistence Economies, edited by Timothy K. Earle, Andrew L. Christenson. New York: Academic Press, 1980.

Hays, T, T.R. Hellier, T.E. Kennerly. Environmental Impact Study of the Elm Fork Region

of the Trini y. Arlington; Department of Biology, University of Texas at Arlington, 1972.

in Economic Prehistory : Studies by members and associates of the British Academy Major Research Project in he Early History of Agriculture; edited by E.S. Higgs. Cambridge: Cambridge University Press, 1972.

Hunt, Eleazer D. “Upgrading Site-Catchment Analyses with the use of GIS: Investigating

the Settlement Patterns of Horticulturalists.” World Archaeology. Vol. 24 (1982): 283-309.

Jarman, M.R. “A Territorial Model for Archaeology.” In: Models in A chaeology; edited

Jarman, M.R., C. Vita-Finzi and E.S. Higgs. “Site Catchment Analysis in Archaeology.”

In: Man, Settlement, and Urbanism. edited by Peter J. Ucko, Ruth Tringham, and G. W. Dimbleby. Cambridge, Mass; Morristown, N.J.: Schenkman Pub. Co.; distributed by General Learning Press, 1972.

Jochim, Michael A. Hunter-Gatherer Subsistence and Settlement: A Predictive Model.

_____.“Breaking Down the System: Recent Ecological Approaches in Archaeology.” In: Advances in Archaeological Method Theory 2. edited by M.B. Schiffer. New York: Academic Press, 1979.

_____.Strategies for Survival: Cultu al Behavior in an Ecological Con ext. New York: Academic Press, 1981.

Johnson, Eileen, and Vance T. Holliday. “The Archaic Record at Lubbock Lake.” In:

Plains Anthropologist 31 (1986): 7-54.

297

Kvamme, K.L. “Geographic Information Systems in Regional Archaeological Research and Data Management.” In: Archaeological Method and Theory 1, Tucson: The

Lynott, M.J. Archaeological Excavations at Lake Lavon 1974. Dallas: Archaeological

Research Program. Contributions in Anthropology No. 16. Southern Methodist University, 1975.

_____.“A Model of Prehistoric Adaptation in Northern Texas.” Plains Anthropologist 26 (92) (1981): 97-110.

Malczewski, Jacek. GIS and Multicriteria Decision Analysis. New York: John Wiley, 1999. McCormick, O.F., R. Filson and J. Darden. The Xerox Lewisville Archaeological Project.

McMahan, C.A., R.G. Frye and K.L. Brown. The Vegetation Types of Texas, Including

Cropland: An illustrated Synopsis to Accompany the Map. Austin: Wildlife Division, 1984.

Research Institute, Inc., 2001.

f

tTexas. Nacogdoches: Stephen F. Austin University, 1974.

University of Arizona Press, 1989. Lebo, S.A. and K.L. Brown. Archaeological Survey of the Lewisville Lake Shoreline,

Denton County, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1990.

Lee, R.B. and Devore, I. Man the Hunter. Chicago: Aldine, 1968.

_____.A Regional Model for Archaeological Research in North-Central Texas. Unpublished Ph.D. Dissertation. Dallas: Southern Methodist University, 1977.

The Institute for Environmental Studies, Denton; North Texas State University, 1975.

McGregor, D.E. and J.E. Bruseth. Hunter Gatherer Adaptations Along the Prairie Margin. Dallas: Richland Creek Technical Series, Vol. III. Archaeology Research Program, Southern Methodist University, 1987.

Minami, Michael. Using ArcMap. : Redlands: ESRI Press, Environmental Systems

Newcomb, W.W., Jr. The Indians o Texas. Austin: University of Texas Press, 1986.

Nixon, Elray S. and R. Larry Willett. Vegetative Analysis of the Floodplain of the Trini y,

Nunley, Parker. An Assessment of Archaeological Resources in the Vicinity of Garza-

Little Elm Reservoir, Richland/Archaeological Society, Dallas: Miscellaneous Papers No. 1, Richland College, 1973.

298

Prewitt, Elton R. Cultural Chronology in Cen ral Texas. Bulletin of the Texas t

Ormsby, Tim, Eileen Napoleon, Robert Burke, Carolyn Groessl and Laura Feaster

Prikryl, Daniel J. A Synthesis of the Prehistory of the Lower Elm Fork of the Trinity. Unpublished Masters Thesis. Austin: University of Texas at Austin, 1987.

_____.Lower Elm Fork Prehistory : a Redefinition of Cultural Concepts and Chronologies along the Trinity, North-Central Texas. Austin: Texas Historical Commission, 1990. Renfrew, Colin. “Book Review: Peter Haggett: Locational Analysis in Human

Geography.” Antiquity 43 (1969): 74.

Press, 1977.

Department of Agriculture, Soil Conservation Service, U.S. Government Printing

Schiffer, M.B. “An Alternative to Morse’s Dalton Settlement Pattern Hypothesis.” Plains

Anthropologist 20 (1975):253-266. _____.Behavioral Archaeology. New York: Academic Press, 1976.

Sciscenti, J.V. Environmental and Cultural Resources within the Trinity Basin. Dallas:

Skinner, S.A. and L. Baird. The Archaeology and History of Lake Ray Roberts, Vol. 1: Cultural Resources Survey. Dallas: Environmental Consultants Inc., 1982a. _____.Archaeology and History of Lake Ray Roberts, Vol 2, Construction Area Testing.

Dallas: Environmental Consultants, Inc., 1982b. _____.The Archaeology and History of Lake Ray Roberts, Vol. 3: Settlement in a

Archaeological Society 52 (1981):65-89.

Getting to Know ARCGIS Desktop, Redlands: ESRI Press: Environmental Systems Research Institute, 2001.

_____.“Alternative Models for Exchange and Spatial Distribution.” In: Exchange Systems in Prehistory, edited by T.K. Earle and J.E. Ericson. New York: Academic

Ressel, Dennis D. Soil Survey of Wise County, Texas. Washington D.C.: U.S.

Office, 1989. Roper, D. "The Method and Theory of Site Catchment Analysis: A Review," In:

Advances in Archaeological Method and Theory 2, edited by M.B. Schiffer. New York: Academic Press, Inc, 1979.

SMU, Institute for the Study of Earth and Man, 1971.

Marginal Zone. Dallas: AR Consultants, Inc., 1985. _____.The Hackberry Creek Archaeological Site. Irving: AR Consultants Inc., 1999.

299

Steward, J.H. “Basin-Plateau Aboriginal Socio-Political Groups.” Bureau of Ame ican r

Urbana, University of Illinois Press, 1955.

Story, D.A. The Archaeology and Bioarchaeology of the Gulf Coastal Plain: Volume I.

Arkansas Archaeological Survey Research Series No. 38., 1990a.

Suhm, D.A. and E.B. Jelks. Handbook of Texas Archaeology: Type Descriptions. Austin:

The Texas Archaeological Society and the Texas Memorial Museum, 1962. Thomas, D.H. “An Empirical Test for Steward's Model of Great Basin Settlement

Tiffany, Joseph A. and L. R. Abbott "Site Catchment Analysis: Applications to Iowa

Texas Press, 1987. Turner, Ellen Sue and Thomas R. Hester. A Field Guide to Stone Artifacts of Texas

Indians. 2nd edition. Lanham: Rowman and Littlefield, 1993.

Ethnology Bulletin No. 120, 1938. _____.Theory of Culture Change; the Methodology of Multi-linear Evolution. Chicago:

_____. The Archaeology and Bioarchaeology of the Gulf Coastal Plain: Volume II. Arkansas Archaeological Survey Research Series No. 38.,1990b.

Patterns.” American Antiquity 38 (1973): 155-176. _____.“Diversity in hunter-gatherer cultural geography.” In: Quantifying Diversity in

Archaeology, edited by Robert D. Leonard and George T. Jones. New York: Cambridge University Press, 1989.

Thünen, Johann Heinrich von. Isolated state; an English edition of Der isolierte Staat.

Translated by Carla M. Wartenberg. edited with an introd. by Peter Hall. Oxford, New York: Pergamon Press, 1966.

Archaeology", Journal of Field Archaeology 9 (1982): 313-322. Tull, Delena. Edible and Useful Plants of Texas and the Southwest. Austin: University of

Vayda, A.P. and B.J. McCay. “New Directions in Ecology and Ecological Anthropology.”

Annual Review of Anthropology 4 (1975): 293-305. Vayda, A.P. and R.A. Rappaport. “Ecology, Cultural and Non-Cultural.” In: Introduction

to Cultural Anthropology, edited by J.A. Clifton. Boston: Houghton, 1968. Verhagen, P. “The use of GIS as a Tool for Modelling Ecological Change and Human

Occupation in the Middle Aguas Valley (S.E. Spain).” In: Interfacing the Past, edited by Kamermans, H. and Fennema, K. Leiden: University of Leiden, Analecta Praehistorica Leidensia, 1996.

Vita-Finzi, C. Archaeological Sites in Their Setting, London: Thames and Hudson, 1978.

300

Vita-Finzi, C. and E.S. Higgs. “Prehistoric Economy in the Mount Carmel Area of Palestine: Site catchment analysis.” In: Proceedings of the Prehistoric Society 36 (1970):1-37.

Webley, D. Soils and Site Location in Prehistoric Palestine. In: Papers in Economic

Prehistory : Studies by members and associates of the British Academy Major Research Project in the Early History of Agriculture, edited by E.S. Higgs. Cambridge: Cambridge University Press, 1972.

Wheatley and Gillings. “Sites,Territories and Distances.” In: Spatial technology and

Archaeology : the archaeological applications of GIS. New York: Taylor & Francis, 2002.

Willmsen, E.N. “Interaction, Spacing Behavior, and the Organization of Hunting Bands.”

Journal of Anthropological Research 29 (1973):11-31. Zubrow, E.B.W. Prehistoric Carrying Capacity: A Model. Menlo Park: Stanford University.

Cummings Publishing Company, 1975.

_____.“Knowledge Representation and Archaeology: A Cognitive Example Using GIS.” In: The Ancient Mind: Elements of Cognitive Archaeology, edited by Renfrew, C. & E.B.W. Zubrow. Cambridge: Cambridge University Press, 1994.

301